Daily numbers, graphs, analysis thread 16

[BigChocFrenzy]

Welcome to thread 16 of the daily updates

Resource links:

Uk dashboard deaths, cases, hospitals, tests - 4 nations, English regions & LAs
MSAO Map of English cases
[[https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/909430/Contain_framework_lower_tier_local_authority__14_August_2020.pdf
Slides & data UK govt pressers
UK added daily by PHE & DHSC
R estimates UK & English regions
PHE Surveillance report infections & watchlists every Thursday
ONS England infection surveillance reports
ONS UK death stats released each Tuesday
ECDC rolling 14-day incidence EEA & UK
Daily ECDC country detail UK
WHO dashboard
Worldometer UK page
Plot FT graphs compare countries deaths, cases, raw / million pop
Covidly.com world summary & graphs
Plot COVID Graphs Our World in Data test positivity etc

We welcome factual, data driven, and civil discussions from all contributors 📈 📉 📊 👍

(Number as requested by @Nellodee - brief answer, 3 if it's not a big institution like a care home.)

1108, 12 deaths, 109 admissions (69 from silly Wales counting and 52 England, so don't actually add up), 60 on ventilators, 186500 tests.
Website still seems unstable, no? A few days to fix is fair, it's getting annoying now.

3? That seems very small. Thank you for finding it for me, though. There is a real joy in tracking down very specific pieces of data from the ether, isn't there?

Sorry, just read in more detail and see that the 3 contacts is possibly the "contacts we haven't already contacted ourselves" number.

I've just tried running the calculations with those figures @MRex. If we use a figure of around 28,000 close contacts a week (allowing a bit for the ones that track and trace miss), and want it to work out so that people with known contacts are 24 times more likely to contract Covid, it works out so that the 66m people without a close contact in the UK have about a 0.0104% chance of catching Covid each week, resulting in 6,930 infections, and the 28,000 people who are close contacts have a 0.25% chance, resulting in 70 infections a week - totalling 7,000 infections a week, which is what I was aiming for.

So this does - as we suspected - mean that only a very small proportion of contacts test positive. Slight warning - the figure of contacts being 24 times more likely to test positive referred just to "contacts", not close contacts. It may be higher for close contacts. Or the figure of 24 times more likely may be based on small numbers and simply be wrong.

Does anyone know why the Wales admissions look so comparatively high compared to both England and their own hospitalisations?

Wales count everyone with suspected Covid in their admissions figures. Rest of the UK only count people when they test positive, so Wales are massively over inflated

Thanks @Firefliess

[quote MRex]I've just read something ridiculous:
www.gov.uk/government/publications/nhs-test-and-trace-statistics-england-methodology/nhs-test-and-trace-statistics-england-methodology
In the Household section it says: "It is likely that cases often advise their household members to self-isolate in advance of these contacts being directly contacted by contact tracers. This results in these contacts not being recorded as reached and asked to self-isolate, which may be a contributing factor to a lower proportion of household contacts being reached in comparison to non-household contacts."
So basically they're making the service percentages look unnecessarily crap by marking people as not contacted who already know - why would they do that? Why not just have a category "already informed and isolating"?

Anyway - 24,197 contacts for 7941 cases = just over 3 each. That excludes complex cases such as care homes and hospitals that may be higher and have super-spread issues:
www.gov.uk/government/publications/nhs-test-and-trace-england-and-coronavirus-testing-uk-statistics-13-august-to-19-august-2020/weekly-statistics-for-nhs-test-and-trace-england-and-coronavirus-testing-uk-13-august-to-19-august[/quote]
Utterly ridiculous and bizarre to make themselves look even less effective than they are. It's a sort of robot response!

@MRex

1108, 12 deaths, 109 admissions (69 from silly Wales counting and 52 England, so don't actually add up), 60 on ventilators, 186500 tests. Website still seems unstable, no? A few days to fix is fair, it's getting annoying now.

That admissions number is still from weds 19th, ignore it! For some reason it doesn't update daily. And the other nation based admissions data is from the 26th, it doesn't update over the weekend I don't think.

Thanks @PatriciaHolm, I could see it was bollocks but no idea why. I like to make allowance a bit of IT glitch when they're rolling out new stuff, but it's been weeks now, any idea if they're actually trying to fix this?

@Firefliess - ah, thank you for helping me think this bit through! The main bit I see as missing in your logic are the ONS-suspected huge numbers of asymptomatic / unknown cases, see my earlier calc that @BigChocFrenzy was also thinking about. (Backed up by Covid-Zoe and the fact of ongoing rumbling transmission.)

Mrex This goes back to the low dispersion factor, K, of COVID
It spreads via clusters

90% of people with COVID infect 0 or 1 person
Superspreaders who are responsioble for the other 10-% can infect dozens or even hundreds

A minority of people shed a huge amount of virus within a certain time frame

I posted earlier on the Studies thread about a detailled RKI analysis of all cases, types of outbreaks and ages

It shows the average number of people infected in a domestic home =3
the average number in a care home outbreak = 19
in a refugee hostel =21

Why do some COVID-19 patients infect many others, whereas most don’t spread the virus at all?

https://www.sciencemag.org/news/2020/05/why-do-some-covid-19-patients-infect-many-others-whereas-most-don-t-spread-virus-alll

SARS-CoV-2 .....Without social distancing, this reproduction number (R) is about three.
But in real life, some people infect many others and others don’t spread the disease at all.
.....
That’s why in addition to R, scientists use a value called the dispersion factor (k),
which describes how much a disease clusters.

The lower k is, the more transmission comes from a small number of people
.....
In a seminal 2005 Nature paper, Lloyd-Smith and co-authors estimated that SARS - in which superspreading played a major role - had a k of 0.16.
The estimated k for MERS, which emerged in 2012, is about 0.25.
In the flu pandemic of 1918, in contrast, the value was about one, indicating that clusters played less of a role.
......
in a recent preprint, Adam Kucharski of LSHTM estimated that k for COVID-19 is as low as 0.1
“Probably about 10% of cases lead to 80% of the spread,”

In Germany, the earliest epicentres e.g. Gangelt, from the Streeck study, were around a few carnivals
Just a few people who were infected by a colleague from China went on to infect hundreds at their regional (outdoor) carnivals and this soon spread through the state

Hence why it has not yet been decided if the (outdoor) Christmas markets will be allowed to go ahead
It has already been decided to cancel Oktoberfest
and the football league games here will continue indefinitely in empty stadiums, no fans allowed

It is impossible to know in advance who will be a superspreader
it is only possible to limit the numbers who can be infected in any one incident

Thank you so much for the explanation @BigChocFrenzy. It suggests even more clearly that big events are a major issue. I'm so surprised by how many countries have allowed big weddings, nightclubs etc.

Coronavirus: Winter plans revealed in leaked Sage report

Warning: Do NOT panic if you hear of this leaked SAGE report elsewhere
It is only a "reasonable worst case" scenario NOT a prediction
Any responsible government has to plan for this, but that does NOT mean it is likely to happen

https://www.bbc.com/news/uk-53954492

A leaked government report suggests a "reasonable worst case scenario" of 85,000 deaths across the UK this winter due to Covid-19.
The document also says while more restrictions could be re-introduced, schools would likely remain open.

But it says the report "is a scenario, not a prediction" and the data are subject to "significant uncertainty".

From that same BBC page it says:

''It also states that by November "policy measures would be put in place to reduce non-household contacts to half of their normal pre March 2020 levels". In other words, restrictions to mitigate the impact - other than school closures - could be put in place.''
So are we likely to be told to reduce social contacts even more than we already are doing?

I’m glad schools are a priority. It’s always up for change but it’s a start.

@MarshaBradyo

I’m glad schools are a priority. It’s always up for change but it’s a start.

Absolutely. No idea how it will work if the teachers are isolating / off sick etc and wondering if DfE were aware of the SAGE report before working out the 4 tier plan in their latest guidance (unless they've updated it today!)

Shit (ha at your name) I’m not sure but perhaps it influenced the latest DfE somewhat. I noticed that quite a few times DfE mentioned keeping schools open as a priority too.

I don’t know how they’ll cope with teachers off particularly in lockdown stage. I suppose Leicester are doing ok? They are lock down still (I think) but not tier two.

Ah yes. I meant if the SAGE report is correct & numbers rise a lot (I'm assuming they'd have to be a massive uptick in cases to achieve anywhere near that amount of deaths!)
So based on that, there could be school closures due to isolating staff etc. If teachers are off sick of course work can't be set by that teacher but I'm hoping the schools have some sort of back up ready to go for children with isolating/sick teachers.

Oh yes I see what you mean. I tend to gloss over predicted numbers these days. But yes if it’s that high I have no idea how that problem will be fixed.

"Reasonable worst case" is a scenario with a small, but not tiny, chance of happening, so maybe 1 in 10, 1 in 20 ?

Thank you @BigChocFrenzy