Coronavirus in the UK - news, lockdown and discussion

[andysays]

Not everyone that is positive develops symptoms though.....that's part of the problem and why they are trialing testing people without symptoms in Liverpool.

That's also true, which is why the whole thing is complicated.

I'm not seeking to tell anyone whether or not they should take a test, in the end it's up to them, but I do think it's worth challenging simplistic or misleading things that others posts, whether around testing or anything else.

 

[IC3D]

The point that you’re not getting is that the tiny number of detected reinfections tells us precisely fuck all about the actual rate. I mean forgive me if I give a bit more weight to someone writing in the lancet than I give to your opinion, but there you go.

For what it’s worth I’m kind of working on similar assumptions, but it’s just daft to go around making definitive statements when the actual researchers are saying ‘honestly we’re not sure’.

Don't know about that. There's a enough data being scrutanised worldwide to indicate a trend.
Cavet: shrugs, in my opinion man.

 

[Supine]

It’s lovely to read a debate about epidemiology conducted by amateurs

I think it's safe to say we've all increased our understanding of epidemiology this year

 

[Orang Utan]

I think it's safe to say we've all increased our understanding of epidemiology this year

Not me! I avoid getting bogged down with stuff I have no understanding of

 

[Cid]

Don't know about that. There's a enough data being scrutanised worldwide to indicate a trend.
Cavet: shrugs, in my opinion man.

From the paper the article I linked is discussing

The degree of protective immunity conferred by infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is currently unknown. As such, the possibility of reinfection with SARS-CoV-2 is not well understood. We describe an investigation of two instances of SARS-CoV-2 infection in the same individual.

From the article:

Due to the paucity of broad testing and surveillance, we do not know how frequently reinfection occurs among individuals who recovered from their first infection.

The CDC position:

In general, reinfection means a person was infected (got sick) once, recovered, and then later became infected again. Based on what we know from similar viruses, some reinfections are expected. We are still learning more about COVID-19. Ongoing COVID-19 studies will help us understand:

• How likely is reinfection
• How often reinfection occurs
• How soon after the first infection can reinfection take place
• How severe are cases of reinfection
• Who might be at higher risk for reinfection
• What reinfection means for a person’s immunity
• If a person is able to spread COVID-19 to other people when reinfected

I could go on... there’s plenty of discussion in various journals etc. All that can really be said is it’s still poorly understood.

 

[Cid]

Not me! I avoid getting bogged down with stuff I have no understanding of

Have you thought of a career in politics?

 

[Orang Utan]

Have you thought of a career in politics?

Lol, fuck no.
(though I can’t be the only person who can’t even begin to start reading about stats and graphs and projections and r numbers, you know, science, lol - it’s just so dry it’s hard to get interested in even if it’s literally a matter of life and death)

 

[Cloo]

Has anyone thought to do any research on surveying where Brits traveled to over summer (UK and abroad), how long and by what means and how many of them developed C19 within a fortnight of return? Would seem to be a useful thing to know ahead of next summer if you could query a decent sample.

 

[Cid]

Lol, fuck no.
(though I can’t be the only person who can’t even begin to start reading about stats and graphs and projections and r numbers, you know, science, lol - it’s just so dry it’s hard to get interested in even if it’s literally a matter of life and death)

Well quite. Shoe-in for health secretary.

 

[elbows]

Why don't you support that idea, when experts who work in this area do? Why should I believe you over all of them?

Your failure to appreciate the nuanced position of various experts in the relevant scientific communities, and your ability to treat science as some kind of monolith when you find certain tentative conclusions appealing and seek to cement them inappropriately as established facts is noted. Nor it is a good idea to mistake a lack of evidence as being strong proof of something, that isnt how scientific evidence works and I repeatedly see people making this mistake and misinterpreting scientific papers as a result. In this case, lack of evidence about reinfection is not a sign that reinfection is a non-issue.

You should not simply believe me. You could use some of what I say as a basis to take a more cautious approach and to explore the topic in more depth. You may also note my track record of shitting on certain assumptions made earlier in this pandemic, inclucing assumptions made by experts who should know better, and how I have been vindicated on those fronts more often than not.

The likes of SAGE choose their language carefully for good reason. Its not just a question of belief, but of strength of confidence in something, and strength of evidence. On this subject the evidence is not sufficiently strong to make really bold claims with yet, so caution is the order of the day when it comes to immunity issues. Which doesnt mean that I have some bizarre and wacky views about immunity that are utterly at odds with the scientific community, far from it, just that its still early days and bold claims and certainties are dangerous.

For example, SAGE meeting 55, 3rd September includes:

There is an antibody response in nearly all infected people, including those who are asymptomatic (high confidence). It is not yet known how long these responses last, the degree of protection conferred, or effects on transmission, but where neutralising antibodies are present protection against infection seems very likely (medium confidence). There is some evidence of antibody levels waning. T cell reactivity may be more widespread in the population than antibodies (medium confidence).

(from https://assets.publishing.service.g...0739_Fifty-fifth_SAGE_meeting_on_COVID-19.pdf )

Medium confidence isnt useless, it means something, but its still a reason to proceed with care and not resort to crude assumptions that may not stand the test of time. Especially as human understanding of the immune system is pretty weak in some areas.

 

[Cid]

There’s a temporary fireworks place near my workshop. It is a) open, b) no-one in it mask wearing, c) busy.

 

[cupid_stunt]

So, the message is back to - Stay at Home > Protect the NHS > Save lives.

But, no one thought of changing the signs on the podiums, that still read - Hands > Face > Space.

 

[Teaboy]

There’s a temporary fireworks place near my workshop. It is a) open, b) no-one in it mask wearing, c) busy.

That's clearly not essential but those pop-up shops are often run by dodgy characters so I guess they are not bothered by rules.

 

[LDC]

NHS head was surprisingly good, and his (single) slide and animation clearly done by someone that knew about information presentation.

 

[Hyperdark]

Loking at the Latest cases/Deaths figures for Wales, I dont think the 17 day firebreak is enough to stem the tide, partly because most people seem to be completely ignoring it with the only difference being the pubs and cafes are shut.
Talking of which, allowing pubs and gyms to open back up at this stage is not going to help

 

[Espresso]

NHS head was surprisingly good, and his (single) slide and animation clearly done by someone that knew about information presentation.

I liked him too. He's obviously used to explaining things plainly to people he knows are not as clever as him, without talking down to them.

If someone could teach that to Johnson and Hancock and Jenrick and Williamson and Gove and all the rest, that would be very nice. I'd suggest that he Simon Stevens could so it, but I'd imagine he's got a lot on his plate at the moment.

 

[Cid]

That's clearly not essential but those pop-up shops are often run by dodgy characters so I guess they are not bothered by rules.

Yeah absolutely... It's more that it's basically in central Sheffield, and I'm pretty sure some police will have driven past it and clearly ignored it. And also that clearly there a bunch of people who are happy to buy their shit.

I'm rather pessimistic for this one. It feels so different from lockdown 1. Popped into Tesco on the way home, goof mask wearing (or at least no worse than normal), but no social distancing and very busy. Though I suppose it is harder to limit customers when it's cold out.

This morning traffic was certainly light, but again nothing like lockdown 1. Loads of fireworks this evening... Just guaranteed that many of those represent the 'little exceptions' attitude. Well. We'll see.

 

[zora]

Just as I was having concerns about Johnson's messaging during the press conference around the deployment of the rapid tests in Liverpool, right on cue, there is an article in the Guardian about the Operation Moonshot trials in Greater Manchester, and its successes or rather failures.
They apparently failed to identify over 50% of cases. Now, sensitivity of these was always going to be lower, and I guess given that you can test vastly higher numbers of people, even picking up 50% of asymptomatic cases that might otherwise go undetected could be somewhat helpful, so maybe they can still have a role to play.
But Johnson in his urge to make things sound better than they are, put an imo dangerous opinion across that those who test negative can act more normalish. Of course, yes, they won't have to self-isolate like the people testing positive but I wouldn't be surprised if people took it as green light to abandon precautions or not take current restrictions as seriously.

 

[Roadkill]

I liked him too. He's obviously used to explaining things plainly to people he knows are not as clever as him, without talking down to them.

If someone could teach that to Johnson and Hancock and Jenrick and Williamson and Gove and all the rest...

I can see one small problem there.

 

[elbows]

Just as I was having concerns about Johnson's messaging during the press conference around the deployment of the rapid tests in Liverpool, right on cue, there is an article in the Guardian about the Operation Moonshot trials in Greater Manchester, and its successes or rather failures.
They apparently failed to identify over 50% of cases. Now, sensitivity of these was always going to be lower, and I guess given that you can test vastly higher numbers of people, even picking up 50% of asymptomatic cases that might otherwise go undetected could be somewhat helpful, so maybe they can still have a role to play.
But Johnson in his urge to make things sound better than they are, put an imo dangerous opinion across that those who test negative can act more normalish. Of course, yes, they won't have to self-isolate like the people testing positive but I wouldn't be surprised if people took it as green light to abandon precautions or not take current restrictions as seriously.

I need to see more data from other trials of the same type of test since they are claiming that the results seen in Greater Manchester dont match other evaluations of that test. I cant judge without the data.

Whats described in that story is especially annoying because this country has been very slow and conservative at using different sorts of tests at all. So the hope would be that we at least get something in exchange for all that was lost by not rolling such things out a lot sooner. That we'd eventually get approved tests with the best accuracy and less chance of making terrible mistakes of using using tests in inappropriate ways.

 

[elbows]

We dont even live in a country that can manage to publish its daily data within 4 hours of the time its supposed to be published. Todays is still absent even now.

"Owing to technical difficulties, we have not received the data for England. We will update the service as soon as possible."

 

[elbows]

Like all good actuaries, I like to do a bit of backtesting of estimates. The 26 October death rate is probably reasonably represented by the 7-day average as at 30 October, which was 237 for the official reported count. So that 230-690 confidence interval for 26 October that we were talking about on 14 October looks like it did encapsulate the true number after all. Sadly, 150 was woefully short.

By the way, I get the impression that the model in question may be the same one that generated the 4000 peak curve that has caused a stink when Whitty & Vallance used it recently to justify lockdown. I cant be 100% sure because the version of the model output that we were talking about that is routinely published gets chopped off at a point not too many weeks into the future. And I dont have time to try to check my facts in other ways. But it seems like a reasonable fit and it does make criticism of the press conference use of it more valid, I'd probably already commented here on this forum about a later run of that model that produced different results, before they even presented that old version.

Covid: Regulator criticises data used to justify lockdown

Government used out-of-date model predicting 4,000 deaths a day in TV briefing.

www.bbc.co.uk

I still havent found time to comment on the most recent version of the model we discussed here last month. So much real data to look at, and my interest in the models decreases at this sort of stage in a wave, that I might not bother. But I'll link to it anyway. And I will still make that final post about how many UK and England deaths were actually reached on October 26th eventually, once its reasonable to assume that most of the data for that date has been reported/published.

https://joshuablake.github.io/public-RTM-reports/iframe.html

 

[cupid_stunt]

Today's figures have finally been published, deaths are 378, up from 280 last Thursday.

 

[souljacker]

Definitely seems like the case rate is levelling off though. Death stats are always behind case stats too. Maybe this 4 week think might just work, at least until January anyway.

 

[elbows]

The weekly surveillance report, which came out on Thursday has this to say about that sort of thing. Although it does lag behind a bit, so its not really able to accurately describe the latest infection picture.

https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/932943/Weekly_Flu_and_COVID-19_report_w45_FINAL.PDF

Detections of COVID-19 cases in England remained high in week 44. Case detections decreased slightly compared to last week though this is likely to be driven by reduced testing over the half term period as well as a lag in results for the most recent days. Overall positivity rates continued to increase. Incidence and positivity rates remain highest in the North of England though there are some indications that positivity is starting to decline in the North East and North West. By age group, cases rates were highest in the 20 to 29 year olds with decreases continuing to be noted in the 10 to 19 year olds. Positivity rates were highest in the 80+ year olds tested through both Pillar 1 (NHS and PHE testing) and in the 10 to 19 year olds tested through Pillar 2 (community testing).

There are quite a number of graphs fromt he report that I would use to make my point, but I shall restrain myself and just post two.

 

[elbows]

Actually forgive me, for I really wanted to include some graphs that show the positivity rates, since the trends have some subtle differences compared to number of cases. And many of these levels are well above the recommended percentage, indicating that testing is not being done at the ideal scale to capture an epidemic of this size properly. I've only focussed on pillar two because otherwise there would be even more graphs.

 

[Supine]

^ when you say above the recommended percentage of 15% - this is considered as 'spreading like wildfire' in the US apparently

 

[elbows]

^ when you say above the recommended percentage of 15% - this is considered as 'spreading like wildfire' in the US apparently

It still drives me mad that when the likes of the BBC reported on some temporary attempts in the USA to limit testing to exclude asymptomatic people, they were prepared to casually mention experts there saying that bit was a vital aspect of controlling the pandemic, and yet that concept never quite seems to make it across the pond when it comes to how they report on our own response and test regime. At least not until our authorities are somewhat ready to actually do it.

But I mention that now because I suppose whether or not you test asymptomatic people would make a difference for how you judge positivity levels.

Meanwhile I cannot let the pandemic pass without recording this particular turd from the Telegraph, a newspaper that has certainly done plenty to demonstrate its shit credentials and priorities in this pandemic. I'm not even interested in the body of the article, I just want to record the headline for posterity.

 

[elbows]

Oh and regarding percentage positive, I think back in May the WHO recommended countries only consider reopening things where it had been below 5% for several weeks first.

 

[Mation]

.