Coronavirus in the UK - news, lockdown and discussion

[elbows]

There is also a hint in there that they've either been asked to comment on the idea of reducing the publics access to certain forms of data, or are concerned the government might try that.

SAGE advises that continuing to provide near real time local information on prevalence would be helpful.

 

[miss direct]

Are masks also not going to be compulsory in medical settings?

 

[Orang Utan]

Are masks also not going to be compulsory in medical settings?

as LynnDoyleCooper testifies that’s already the case

 

[miss direct]

Ugh. Just had my nurse housemate tell me how happy she is not to have to wear one anymore. Seems irresponsible.

 

[kalidarkone]

Ugh. Just had my nurse housemate tell me how happy she is not to have to wear one anymore. Seems irresponsible.

FFS!

 

[elbows]

I dont think I'll talk about the modelling too much today, I'm sure there will be plenty of other opportunities to do so.

But here are the links for anyone interested in that much detail.

If forced to summarise, most model results are better than older modelling exercises, eg less hospitalisations. Partly due to the delay to step 4, partly due to improved estimates regarding vaccine effectiveness etc. But there is still wide variation, and some of the parameters are very sensitive. So a lot of these documents end up being a lesson in the uncertainties. eg they dont know how many unvaccinated people there are because population estimates are far from perfect. And rather unsurprisingly all the results depend on how quickly people return to pre-pandemic behaviours, and they dont know what the reality of this will be. Hence the emphasis on going slowly in todays press conference.

SPI-M-O: Summary of further modelling of easing restrictions – Roadmap Step 4 on 19 July 2021, 7 July 2021

Imperial College London: Evaluating the roadmap out of lockdown for England: modelling the delayed Step 4 of the roadmap in the context of the Delta variant, 7 July 2021

JUNIPER: Transitioning from non-pharmaceutical interventions to vaccination to control COVID-19 transmission, 7 July 2021

LSHTM: Updated roadmap assessment – prior to delayed Step 4, 7 July 2021

University of Warwick: Roadmap scenarios and sensitivity – Step 4, 6 July 2021

 

[kabbes]

I dont think I'll talk about the modelling too much today, I'm sure there will be plenty of other opportunities to do so.

But here are the links for anyone interested in that much detail.

If forced to summarise, most model results are better than older modelling exercises, eg less hospitalisations. Partly due to the delay to step 4, partly due to improved estimates regarding vaccine effectiveness etc. But there is still wide variation, and some of the parameters are very sensitive. So a lot of these documents end up being a lesson in the uncertainties. eg they dont know how many unvaccinated people there are because population estimates are far from perfect. And rather unsurprisingly all the results depend on how quickly people return to pre-pandemic behaviours, and they dont know what the reality of this will be. Hence the emphasis on going slowly in todays press conference.

SPI-M-O: Summary of further modelling of easing restrictions – Roadmap Step 4 on 19 July 2021, 7 July 2021

Imperial College London: Evaluating the roadmap out of lockdown for England: modelling the delayed Step 4 of the roadmap in the context of the Delta variant, 7 July 2021

JUNIPER: Transitioning from non-pharmaceutical interventions to vaccination to control COVID-19 transmission, 7 July 2021

LSHTM: Updated roadmap assessment – prior to delayed Step 4, 7 July 2021

University of Warwick: Roadmap scenarios and sensitivity – Step 4, 6 July 2021

I just had a look at the Warwick one. Basically (assuming my quick read is right), hospitalisations peaking a bit lower than wave 1 rather than just higher than wave 2 (as would have been the case with a June opening). So still fucked, just slightly less so.

 

[elbows]

I just had a look at the Warwick one. Basically (assuming my quick read is right), hospitalisations peaking a bit lower than wave 1 rather than just higher than wave 2 (as would have been the case with a June opening). So still fucked, just slightly less so.

That is a fair description of one section and graph from the Warwick paper, I would agree.

But these exercises are riddled with all sorts of graphs where they demonstrate what the effects of changing different parameters are. So there is no single graph or story of the wave which I could possibly settle on as being a description of what they think will happen. Uncertainty abounds and we have to consider all these ranges and variables, at which point we end up with a rather broad and mushy picture of what is in store.

In this particular case the thing you zoomed in on seems to be a reasonable way of considering what sort of difference the step 4 delay may have meant. Note also the different timing, with the peak of hospital admissions in that particular graph occurring towards the end of August. But Im far from convinced its the best graph to give us a guide as to what this wave will actually look like, and its timing.

When seeking to somewhat capture the mushy range of possibilities, I suppose I recommend reading the SPI-M-O summary document rather than zooming in on one particular models output. Still need to zoom in as well, for example they point out that if they merge all the different models results, the timing of the peak goes all fuzzy and broad rather than sharp, which I dont think they expect to be a fair reflection of what will actually happen.

Having said all that I still have to pick some graphs in particular when I want to talk about this stuff here, I will have to zoom in to get to the detail and the nature of the uncertainties. But I havent read all of the modelling documents yet and I will want to sleep on it before deciding which graphs are most suitable for this purpose.

 

[LDC]

as LynnDoyleCooper testifies that’s already the case

I'm not sure tbh.

I think they seem not to be now a legal requirement even in healthcare settings, but my work has emailed us saying nothing will change re: pandemic stuff after the 19th, and we 're not to expect it to at any point in the near future. The email was good, mentioned that the government don't seem to be following best science etc.

E2A: Ah, maybe Orang Utan was making a subtle point about my lovely anti-mask co-worker?! Anyway, my comment above was more about whether they are or not required in healthcare. I expect they will stay across the board tbh, at last for staff. Enforcing them for the public might be much harder though.

 

[Orang Utan]

i didn’t think it was subtle but a direct reference, LynnDoyleCooper !

 

[elbows]

Oh in terms of the North East 'variant signal' story that came up here, I previously pointed out the limitations of the language used in PHEs denial.

I have since seen this which is a lot closer to explicitly denying that a signal has been spotted in that region. They dont use the word signal but they've not left much room for that possibility to remain undenied, in that if the rise in that region is consistent with the rise in Delta, that implies no signal.

My mind inevitably remains somewhat open to various possibilities but I've not had any sort of moment with this story yet where I thought a special and alarming new explanation for the North Easts numbers was really required.

 

[William of Walworth]

This excellent article by "Anonymous Doctor" in yesterday's Observer, is pretty informative (IMO) about potentially horrendous consequences for hospitals, particularly A & E departments and Intensive Care units :

I'm an intensive care doctor. Covid patients are getting younger this time

One key aspect of the current policy has not been discussed at all. The vaccine appears to have broken the link between infections and deaths, which is fantastic. But there are 5.3 million people with non-Covid related diseases waiting for treatment. If Covid infections are left to run rampant, with no policy measures to slow them down, how on earth is the NHS going to address the backlog?
The first thing that will suffer is non-urgent operations. We can’t turn away people needing emergency surgery, so we have to have the beds for them. Elective surgery requires critical care capacity, so when something has to give it is inevitably going to be that. That’s devastating, given the backlog of people waiting for treatment.
As infections rise, so too will admissions. As the hospital beds fill up with Covid patients, we will have to stop other work.

(ETA : Correction : The above article was in fact in The Guardian for Monday 12/7 )

 

[andysays]

No, so I suppose if anyone catches Covid there in the next few weeks and then dies, it'll all be my fault.

I was just wondering how the management might have responded to you bringing these perfectly reasonable concerns to their attention, TBH.

Now I guess we'll never know.

 

[platinumsage]

Yes and there was even a bit of 'maybe you want to get a jab because you want to go on holiday' in there somewhere today.

This is pretty impressive in regard to waving the idea of vaccine passports around:

 

[Cid]

This is pretty impressive in regard to waving the idea of vaccine passports around:

When did he announce vaccine passports for cafes?

 

[gentlegreen]

When did he announce vaccine passports for cafes?

 

[bimble]

If he'd said proof of vaccination OR a totally useless dead easy to fake negative test result then what would have happened idk. Interesting.
Maybe i'm wrong and things like this could convince a lot of people to book jabs who otherwise won't.

France reports rush for vaccines after Macron tightens Covid rules

More than 20,000 people a minute book jabs after cafes, malls and trains ruled out of bounds for the unvaccinated

www.theguardian.com

eta oh its the same thing, he said vaccination OR a negative test result. Do they have the same sort of fairly useless lateral flow tests as we have?

 

[Cid]

View attachment 278354

Ahh, the end of the graph cropped on my phone.

 

[two sheds]

Yes I spent a fruitless minute looking, too.

 

[BCBlues]

Yes I spent a fruitless minute looking, too.

Same here

 

[quimcunx]

Ahh, the end of the graph cropped on my phone.

It's also cropped on my extra wide work monitor.

 

[BillRiver]

It's also cropped on my extra wide work monitor.

I saw it but only by clicking on the tweet itself.

 

[cupid_stunt]

If he'd said proof of vaccination OR a totally useless dead easy to fake negative test result then what would have happened idk. Interesting.
Maybe i'm wrong and things like this could convince a lot of people to book jabs who otherwise won't.

France reports rush for vaccines after Macron tightens Covid rules

More than 20,000 people a minute book jabs after cafes, malls and trains ruled out of bounds for the unvaccinated

www.theguardian.com

eta oh its the same thing, he said vaccination OR a negative test result. Do they have the same sort of fairly useless lateral flow tests as we have?

Yes, they will allow negative tests, but they'll have to pay from them, from Sept.

The same requirement will be extended to bars, cafes, restaurants, shopping centres, hospitals, long-distance trains and planes from 1 August, the president said.
Free Covid tests, meanwhile, will end in September, “to encourage vaccination”.

 

[elbows]

Yup. Since day one of this the government's power move has been positioning themself so that they can blame the public when shit inevitably gets worse.

Its totally understandable that this complaint and suspicion has always come up.

I dont think its a proper reflection of all the forces at work though - the fact is that the pandemic has always depended on how the masses behave. But the government buck passing doesnt really work in this way in practice because ultimately government policy and rhetoric gets blamed for how people behave. Plus governments are told by their experts to focus on encouraging words, reinforcing the desired behaviours and praising the public for their pandemic behaviour.

This stuff is bound to come up again more now because the modelling makes it very clear that there is a wide range of possibilities for this wave and when the modellers adjust thing like how quickly people return to pre-pandemic behaviour, they get massively different wave sizes.

If the chosen approach goes wrong then the government will get most of the blame, whatever our other fears about the blame game may be.

 

[elbows]

For example here is the simplified state broadcaster framed description of ths current situation and the modelling. Its Nick Triggle again. Last time I mentioned him it led to me ranting about what the BBC output would be like if a nuclear power station melted, and now there is a nuclear analogy in this article.

Covid: Will 19 July unlocking gamble pay off or backfire?

Lifting remaining restrictions will come at a price, but could it help us finally get on top of Covid?

www.bbc.co.uk

One expert likens what is being done to taking the "control rods out of a nuclear reactor".

"This is going to make things worse than they are now and we don't know when it will peak," says Loughborough University data analyst Dr Duncan Robertson.

I dont think I have time to start fishing out various graphs and figures from the modelling papers today. This simplified boiling down of the Warwick modelling to two graphs in the above article will have to do for now. I dont think it quite captures the full range of possibilities but its a start.

 

[elbows]

Only other thin I'll say about all the recent modelling is that I dont think it begins to cover all the possible scenarios. In particular I dont want to exclude the possibility of an earlier peak than shown in the vast majority of the graphs. This doesnt mean I've suddenly turned into an optimist, but there are so many variables including a potentially wider range of human behavioural responses than they've modelled for, such as more people going 'fuck this' and changing behaviours in the other direction, back away from the old normal rather than just different speeds of still heading towards that old normal. I also dont relaly know if they've modelled properly all the potential dynamics that come from the gloomy mood music, the heightened sense that this plan is stupid, the full effects of lots of positive cases leading to disruption and less mixing oppotunities etc.

 

[Wilf]

There's certainly a sense of there being a phoney war at the moment. We might see a Euros blip in the daily cases next week, but otherwise it all hinges on what follows from the opening up. Not just the (inevitable) rise in cases but the speed with which that happens. Government are a lot less bullish than they were about it even 5 days ago.

 

[elbows]

Yeah and what happened in the Netherlands probably hasnt helped their nerves. In terms of SAGE advice and the modelling, its been about a week now since they had these reports behind the scenes.

When reading all the modelling documents its quite clear that the experts dont know the exact impact of seasonality, they dont quite know how many people have been infected in the past, they dont know how many unvaccinated people there are because the population estimates arent brilliant. Combine those things with the unknowns about behaviour and the uncertainties are greater than they've ever been, especially without the certainties that lockdown brings to peak timing.

For example these are the sorts of modelling graphs I'd be diving into properly here if I had more time. In this case the dotted and dashed lines shows what happens to hospital admissions in their model if the population is larger than has been assumed, and therefore vaccine uptake so far would have been lower than estimated. And each different graph in a different colour reflects a differnt sort of return by the public to pre-pandemic behaviours. Things like how abruptly there is a jump part of the way back to old behaviours, and then how long it takes for the rest of the return to normal to happen.


From the Warwick modelling document which also includes a bunch of very similar charts where instead of looking at different potential numbers of unvaccinated people, look at different seasonality, different future vaccine uptake in younger age groups, different speed of vaccine rollout.

https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1001172/S1302_University_of_Warwick_Road_Map_Scenarios_and_Sensitivity_Step_4.2__6_July_2021__1_.pdf

Just had time to squeeze this post out but no more from me on the modelling today unless I find some time this evening.

PS. in terms of what those different colour scenarios actually mean, I've stuck the relevant charts beind a spoiler tag. Four scenarios are where there only a gradual change to behaviour over time, with differences in speed, and 4 where there is an initial leap of behavioural changes on the 19th and then gradual change on top over subsequent months. I dont really consider scenarios where almost everyone goes back to pre-pandemic behaviour very quickly to be plausible but its still interesting to see the results.

Spoiler

 

[elbows]

The other thing to keep in mind about the current UK plan is that it was partly always just going to be a political trick - moving away from legislation that the tory backbench loons and certain newspapers go mad about, but in practice still having quite a few of those rules in place in practice via stuff like the dreaded 'corporate responsibility'. Its an incredibly dangerous game with real implications, so I'm not trying to claim that the level of adherence seen before will remain intact, but its certainly far short of the freedom day bullshit they tried to pretend it was. The changes would have been more real and dramatic if it werent for Delta, so now we have more fudge on our hands.

The FT noticed this and stuck it on their front page today.

 

[gentlegreen]

The neighbour opposite has COPD and has been locked-in since last March - except when she was carted off to hospital.
A workman has turned up - probably to put in a wheelchair ramp and is maskless as well as showing his pants...