Coronavirus (2021) thread

[Healdplace]

ZOE APP DATA

TODAYS DATA

REGIONS:-

MOST REGIONS UP ON ZOE AGAIN BUT BY ONLY SMALL NUMBERS




North West RISES FROM 745 / 972 TO 764 / 996 - STILL FIFTH of the 12 regions.

But looks set to top 1000 again tomorrow.


Scotland IS DOWN to second with WALES now top.

North East is back up to third UP from 698 / 1091 TO 710 / 1110

With Yorkshire down to fourth



East Midlands falls back into the lower watch zone.


DOWN FROM 637 / 881 TO 629 / 875.

South West (which has been plummeting day by day in the real cases all week - as the regional data each evening on here shows- is now the LOWEST region ahead of London

FALLS FROM 473 / 625 TO 468 / 622

SOUTH WEST HAS BY FAR FALLEN THE MOST IN THE REAL CASES TOO



West Midlands though is UP on 527 / 754 FROM 511 / 733, South East (DOWN to 521 / 640 FROM 524 / 642) and East (DOWN to 492 / 651 FROM 511/ 670).

THE SOUTHERN REGIONS AS IN THE REAL CASE DATA FALLING ON ZOE WHILST THE NORTH IS EDGING UP AGAIN.

UP TOP:-

WALES UP FROM 762 / 1083 TO 795 / 1125


Scotland (2nd) is UP again on 814 / 1090 FROM 805 / 1077


Yorkshire third is DOWN on 755 / 1024 FROM 764 / 1032


N Ireland also DOWN on 292 / 1056 FROM 301 / 1064

 

[roubaixtuesday]

It is the impact of the vaccines on the length of stay in hospital - reduced to a few days on average not weeks

I agree with the rest of your post, but I don't think this is accurate - there doesn't seem AFAICT any reliable data on hospital stay lengths, but the amateur analyses around seem to suggest there's not been a massive shift throughout the pandemic, and certainly not from weeks to days. Around 8 days seems to be where we are.

I think the big impact is likelihood of getting into hospital in the first place, not how long you end up there

 

[roubaixtuesday]

I take huge issue with Sage and the predictions they make. So far their doom predictions have not even come close to fruition

Initial projection 250-500k deaths without mitigation.

Outcome until vaccination: 150k deaths with extensive mitigations.

 

[roubaixtuesday]

If I said to my boss my sales figures for this month are between 0 and 1,000,000 he would say OK, there's the door, thanks for your service. I think the SAGE analysts are unaccountable and covering their own arses, they leave the 0.000001% scenario open so they can say, we warned you and you didn't prepare.

I think this criticism is misguided, for two reasons:

1) They publish, in some detail, all of their assumptions and models. Multiple different groups using different models are used to ensure there's not one methodology being relied on. They're highly transparent.

2) Models are not intended to produce precise predictions. COVID is inherently impossible to model precisely over long periods:
- we don't know what level of immunity exactly is conferred by infection or vaccination and how it changes over time.
- how people behave has a dramatic effect on transmission and is unpredictable
- how people behave is influenced by their perception of risk, so there's a feedback loop
- the mathematics of exponential growth mean that really small differences in the above make huge differences in the prevalence at any time.

So, now we're beyond the very binary "either unmitigated or locked down" phase of the pandemic, all modelling can do is inform the spread of potential outcomes, and something about their likelihood.

The problem with models is that people use the outputs for purposes other than what they're intended for.

 

[bobbyowenquiff]

That's my understanding as well. Case numbers in England have not jumped 3 to 10 fold in the past week or so and don't look likely to do so in the next week or two? the numbers reported are the headline on the BBC website this morning? These predictions look like they are wildly off the mark and come across as total scaremongering when provided with no explanation or justification? Why are none of the professional journalists at these briefings asking such questions about things like this?

The media has lurched from one extreme to another since day one. Some outlets love spreading doom gloom and panic others take a rigid anti-lockdown stance whatever the situation. Very few take the sensible accurate centre ground. Very few stories explain the big picture and the wider context. The stats are twisted and distorted to suit preconceived angles. With a few exceptions the UK press has been pathetic in the way it has covered Covid. The questions at the daily conference are mostly embarrassing and totally politicised.

 

[Healdplace]

ZOE APP DATA


TODAY


GREATER MANCHESTER:-


Bury - has fallen but still the highest in GM . UP on 20, 996 FROM 22, 230 - one of the highest in the North West.

A surprise is to see Manchester now in second UP on 13, 393 FROM 11, 185

Stockport is now in third place and is DOWN very slightly again on 12, 835 FROM 13, 330

Oldham is up to third though DOWN on 12, 110 FROM 12, 992

Tameside down to fifth place and matching its better real numbers is DOWN by more on 12, 038 FROM 13, 149

THE OTHER FIVE BOROUGHS ARE ALL NOW IN THE LOWER WATCH ZONE


Trafford is in sixth and UP on 8374 FROM 7912

Rochdale in seventh is UP on 7754 FROM 5825

Wigan in eighth is UP on 7543 FROM 6706

Salford is DOWN big in ninth as in the real data on 7171 FROM 11, 290




BUT BEST IN GM IS STILL:-

Bolton DOWN again on 3889 FROM 3989

One of the lowest numbers in GM In many weeks







REST OF NORTH WEST


West Lancashire - the region between Bootle, Formby and Southport in west and Wigan is UP on 11, 234


Liverpool DOWN on 13, 456

Sefton up on 14, 176

AND St Helens back up on 15, 794

also



Chorley is UP again too on 17, 803 FROM 17, 765

As in the Lake District DOWN quite a bit to 15, 479


Cheshire East - is DOWN again and out of the higher watch zone on 9864.

Still much better than Chester and Cheshire West - though they are down on 13, 781 FROM 15, 924

Warrington is UP on 12, 110

 

[bobbyowenquiff]

Listening to the news it seems to be covid and seasonal flu cases together which is what they are talking about when mentioning the large hospitalisation figures. There does seem to be some concern re flu this year as there were very few cases last year or in other parts of the world on which to base this years vaccine.

The usual bed occupancy rate in winter in the NHS is apparently between 90 and 95 per cent so it doesn't take much to tip it over the edge. Perhaps what we are seeing is more of a beds crisis than a Covid crisis, not that anyone would dare to ever criticise the "best health service in the world."

 

[Healdplace]

I agree with the rest of your post, but I don't think this is accurate - there doesn't seem AFAICT any reliable data on hospital stay lengths, but the amateur analyses around seem to suggest there's not been a massive shift throughout the pandemic, and certainly not from weeks to days. Around 8 days seems to be where we are.

I think the big impact is likelihood of getting into hospital in the first place, not how long you end up there

I trust you on the data. I was basing this on several comments on the news over past weeks from NHS staff saying they were finding patients were staying several days less on average than last Winter.

I have also noticed how often - it happened this Monday - a big influx of patients comes in and a day or two later a lot of them suddenly go out again. They might not be the same patients but it creates an impression of shorter stays.

Plus there are 800 or so daily going in in England every day but the numbers week to week going up by only 100 as of last night. Does not suggest lengthy stays for many.

 

[roubaixtuesday]

I trust you on the data. I was basing this on several comments on the news over past weeks from NHS staff saying they were finding patients were staying several days less on average than last Winter.

I have also noticed how often - it happened this Monday - a big influx of patients comes in and a day or two later a lot of them suddenly go out again. They might not be the same patients but it creates an impression of shorter stays.

Very, very roughly, the length of stay should be (inpatients)/(patients admitted per day).

Here's that plotted direct from the govt dashboard data



(the patients data is lagged by an arbitrary 8 days otherwise you get the effect of rapidly changing numbers)

This is *not* any kind of definitive analysis, could be skewed in all sorts of ways and certainly shouldn't be relied on, but you get the general picture - the evidence for radical changes in hospital stay length is lacking.

 

[Healdplace]

WALES DATA

10 deaths - was 10 last week

2298 cases - was 2669 last week - another small fall here again.

10.2% positivity - was 12.5% last week

500 patients - was 428 last week

48 ventilated - was 41 last week