Coronavirus - General Chat - No statistics

[jfgw]

XFool wrote:

Nearly a third of English hospital trusts exceed first peak of Covid patients

I just don't buy this. Despite the normal cautions over behavioural changes this is certainly not borne out by

https://assets.publishing.service.gov.u ... 20wk47.pdf
Eb

"Nearly a third of English hospital trusts exceed first peak of Covid patients" implies that more than two-thirds of English hospital trusts do not exceed the first peak of Covid patients. There is a lot of regional variation.


Julian F. G. W.

[XFool]

XFool wrote:

Nearly a third of English hospital trusts exceed first peak of Covid patients

I just don't buy this. Despite the normal cautions over behavioural changes this is certainly not borne out by

https://assets.publishing.service.gov.u ... 20wk47.pdf

The latest bulletin.

So where is the problem? Cases are now going down, this is being reported in the current news - could lockdown actually be working?
Doubtless, 'alternative', more complicated, explanations are available. They usually are.

PCR testing is well known for predicting pseudo-epidemics by false positive testing (the experience of Swine Flu being one well known example). I would be very interested to know how statisticians would interpret following the LF testing in Liverpool to the earlier date from that city.

As I already posted: You don't get admitted to hospital with a case of the false positives.

[johnhemming]

Although people do not get admitted to hospital with a false positive they may be tested positive falsely whilst in hospital for something else although that will only be really small numbers and does not IMO affect the figures materially.

My view about this particular admissions curve is that it is likely to be flatter, but more extended than the previous. Hence although the number of admissions has peaked there will still be material numbers of people admitted for some time. Hence it may be that there is more pressure on numbers being cared for. Additionally whereas London, with its high level of prior infections, is not seeing that many people admitted as a proportion of what happened previously other regions are. Particularly the South West (which may be mainly around Avon, but we don't have detailed figures) where the peak admissions is around the same as earlier in the year.

The NE got to 92% of the previous peak, but is on its way down now. (although total numbers in hospital may still increase depending upon discharge rates).

[XFool]

Latest information:

Covid-19 in the UK: How many coronavirus cases are there in your area?

BBC News

[dealtn]

XFool wrote:

Nearly a third of English hospital trusts exceed first peak of Covid patients

I just don't buy this. Despite the normal cautions over behavioural changes this is certainly not borne out by

https://assets.publishing.service.gov.u ... 20wk47.pdf
Eb

"Nearly a third of English hospital trusts exceed first peak of Covid patients" implies that more than two-thirds of English hospital trusts do not exceed the first peak of Covid patients. There is a lot of regional variation.


Julian F. G. W.

Yes, but "Over 2/3rds of Areas Below Covid Peak" doesn't make attractive enough headlines does it?

[johnhemming]

Interestingly I have found a link to reasonably up to date figures for hospitalisations in Switzerland (inter alia)
https://www.swissinfo.ch/eng/swiss-stat ... -/45674308

They don't give the daily figure, but instead give the numbers in hospital. The peak rate is around the end of October start of November.
Switzerland decided to have increased restrictions, but no autumn lockdown.

[johnhemming]

I have also found the same information for Italy
https://www.statista.com/statistics/112 ... eak-italy/

[johnhemming]

I have also found the daily hospitalisations for Belgium
https://www.statista.com/statistics/110 ... n-belgium/

[johnhemming]

This is an eu modelling document which also includes some information about restrictions
https://www.ecdc.europa.eu/sites/defaul ... r-2020.pdf

[XFool]

"Nearly a third of English hospital trusts exceed first peak of Covid patients" implies that more than two-thirds of English hospital trusts do not exceed the first peak of Covid patients. There is a lot of regional variation.

Yes, but "Over 2/3rds of Areas Below Covid Peak" doesn't make attractive enough headlines does it?

Nobody seems to be mentioning the sub-heading:

"Scientists warn that scrapping or relaxing tier system too quickly could imperil NHS"

As ever, it's the dynamics of the situation that matters - that'll be what "scientists" attend to - not just a flash photo of the latest stats.

[dealtn]

"Nearly a third of English hospital trusts exceed first peak of Covid patients" implies that more than two-thirds of English hospital trusts do not exceed the first peak of Covid patients. There is a lot of regional variation.

Yes, but "Over 2/3rds of Areas Below Covid Peak" doesn't make attractive enough headlines does it?

Nobody seems to be mentioning the sub-heading:

"Scientists warn that scrapping or relaxing tier system too quickly could imperil NHS"

As ever, it's the dynamics of the situation that matters - that'll be what "scientists" attend to - not just a flash photo of the latest stats.

The headline writer's favourite weasel word "could"! I haven't read the article, but it's fair to say a headline writers job is far removed from being objective. Hopefully the article, which should have a different objective is more, well objective.

[johnhemming]

SAGE need to maintain restrictions to justify the previous restrictions.

[XFool]

Yes, but "Over 2/3rds of Areas Below Covid Peak" doesn't make attractive enough headlines does it?

Nobody seems to be mentioning the sub-heading:

"Scientists warn that scrapping or relaxing tier system too quickly could imperil NHS"

As ever, it's the dynamics of the situation that matters - that'll be what "scientists" attend to - not just a flash photo of the latest stats.

The headline writer's favourite weasel word "could"! I haven't read the article, but it's fair to say a headline writers job is far removed from being objective. Hopefully the article, which should have a different objective is more, well objective.

Perhaps it is more important to attend closely to what the sub-heading is actually referring to?

Scientists warn that scrapping or relaxing tier system too quickly could imperil NHS

Details can be found within the article.
e.g.
Openshaw [a professor of experimental medicine at Imperial College London] said: “We scientists are very concerned indeed about relaxation of precautions at this stage. The rates are still too high, there’s too many cases coming into hospitals, too many people dying. And if we take the brakes off at this stage, just when the end is in sight, I think we would be making a huge mistake,” he told the BBC.

[Lootman]

Cases are now going down, this is being reported in the current news - could lockdown actually be working?

Could be, but then the curve had turned down before the lockdown started. So how would you know if it would not have declined further anyway?

[XFool]

SAGE need to maintain restrictions to justify the previous restrictions.

Or the purveyors of 'alternative' explanations need to continue attempting to discredit* reality? Since reality is doing a pretty good job of discrediting their Summer prognostications: "The pandemic is all over, there will be no second wave..."

* There is no pandemic: It is all just false positive PCR tests; It isn't a second wave because: a) It's a seasonal wave b) It's the wrong shape of wave c) It's not as big as the first wave, so it doesn't count.

[redsturgeon]

Cases are now going down, this is being reported in the current news - could lockdown actually be working?

Could be, but then the curve had turned down before the lockdown started. So how would you know if it would not have declined further anyway?

https://ourworldindata.org/coronavirus- ... rSort=desc

Have a look at what happens when you don't lock down

John

[XFool]

Cases are now going down, this is being reported in the current news - could lockdown actually be working?

Could be, but then the curve had turned down before the lockdown started.

Far from obvious to me.

So how would you know if it would not have declined further anyway?

Sometimes seems to me these pesky PCR etc. tests "work" when required and then "not work", also when required! Perhaps they are argument dependent tests?

[Lootman]

Cases are now going down, this is being reported in the current news - could lockdown actually be working?

Could be, but then the curve had turned down before the lockdown started.

Far from obvious to me.

The plots I have seen show cases declining from late October, a week or two before lockdown started. May vary by area of course, but it constitutes reasonable doubt about whether the lockdown was necessary, at least nationally.

[zico]

SAGE need to maintain restrictions to justify the previous restrictions.

Or the purveyors of 'alternative' explanations need to continue attempting to discredit* reality? Since reality is doing a pretty good job of discrediting their Summer prognostications: "The pandemic is all over, there will be no second wave..."

* There is no pandemic: It is all just false positive PCR tests; It isn't a second wave because: a) It's a seasonal wave b) It's the wrong shape of wave c) It's not as big as the first wave, so it doesn't count.

Don't forget 'This second wave will be confined to younger people, so deaths will be much lower. "

[XFool]

The plots I have seen show cases declining from late October, a week or two before lockdown started. May vary by area of course, but it constitutes reasonable doubt about whether the lockdown was necessary, at least nationally.

That may well be true, about different areas, but the previous arrangements were deemed not to be working well. Already the Tier system for after the lockdown seems to be proving not 100% popular!