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 📈 📉 📊 👍

Does anyone know how they do the random samples for the PHE reports. Is it literally random numbers or do they make some attempt to make sure every region has similar representation?
Also wondering if the returns good - I remember hearing about cash/voucher incentives for one report which I guess could help.
Random thoughts! Maybe no answers.

I'm on the ONS Covid study, our address was selected at random.
Incentives are good, £50 vouchers per person for the first swab tests, then £25 vouchers for subsequent ones. Once a week for the first month then once a month for a year.
They test the swabs for infection and for antibodies but we'll only be told results if Covid positive, and not at all for antibodies.

[quote YoshimisMum]@EducatingArti
Salford number 29 on here[/quote]
Hello @YoshimisMum

Where is this table from please? I'd like to see this

@sunseekin - DHSC used Ipsos MORI to get the random representative sample from the NHS database of everyone registered woods over the age of 5 to send out letters; this is info from the pillar 4 overview because I found that one first, but they were used with ONS too (not Imperial though, that one's UCL):
www.thelancet.com/journals/landia/article/PIIS2213-8587(20)30268-0/fulltext.

It's not like Ipsos Mori have ever got election nor referendum results wrong . They have all the usual demographic info to calculate what counts as representative and pick random individuals within that, so it's clearly been set up with the aim of not being biased towards any specific regions / age groups etc.

Ignore the random "woods" in that, my phone's auto-incorrect is desperately trying to communicate secret messages to the outside world (that it's being over-used?).

More numerical musings, please excuse the over-posting today. I read that back in July the swab test found people who had been a close contact were 24 times more likely than the general population to catch covid. This was with infections at around 48,000 or 1 in 1400. Now, dividing 1400 by 24 gives 58. So only a 1 in 58 chance of catching covid even when identified as close contact. How can that be right? How can that even align to an R of 1? What have I missed? Thanks

I suppose either people must have, on average, 58 contacts for an R of 1, or that R must be made up of people outside close contacts?

I'm not sure you maths is quite right there @MRex. I think you'd need to know how many close contacts there are each week (call this number c), how many other people who are not known to be a close contact (population of UK, minus c) and then use simultaneous equations to work out what proportion of close contacts are actually positive. I don't know how many close contacts there are each week, so can't work this out - though presumably you could hazard a guess from the track and trace reporting?

I would imagine it's quite a low proportion of contacts are positive though, as there was research out early on that found that even within households the proportion of people who caught it was only around 10-25%, which seems very low and suggests to me that a small proportion of people are natural super-spreaders who spread it to dozens, whilst the majority are hardly infectious.

[quote YoshimisMum]@EducatingArti
Salford number 29 on here[/quote]
Is there are full list of LA’s anywhere that you know of so I can see where my LA is?

@Augustbreeze, oh yes, lovely Murray, I hope she's doing OK. We lost my Mum in June (not Covid) and I thought often of the wonderful carehome staff like Murray who give their all (and beyond) to their residents. I couldn't be there but somebody in the carehome was.

Does anyone know how to see any estimates of how many people have had it in total, so in addition to actual positive tests all the people who will have had it but not had a test.

Thanks @Firefliess. You're right that the answer must be the super-spreaders, leaving lots of people who don't infect anyone. Even if 10% of people result in 80% transmission then those 10% don't have to transmit to all that many people really with those numbers, they just have to be at the right infection stage and talking.

We have just had a letter inviting us to join the ONS study! I have registered us today - very excited to be able to do some small thing to help.

I had no idea until I got the letter that they gave you vouchers for participating! We would have done it anyway for nothing so that’s just a bonus! @InMySpareTime out of interest what can you use the vouchers for??

@Farlow - you can look up your area on the surveillance spreadsheet figure 10, it's the 5th document down. If your area is on the watchlist then look at the 4th document for more info:
www.gov.uk/government/publications/national-covid-19-surveillance-reports.

@Farlow
ourworldindata.org/covid-models

Loads of places! They're Sodexo e-vouchers, sent by email a couple of days after each visit.

And here:

I'd be interested in seeing some peripheral tracking data - things like, how many close contacts on average does each person refer to contact tracing?

@MRex @CoffeeandCroissant that’s great. Thank you both.

Which reminds me, we haven't seen the Google mobility data for quite some time.

From baseline, for the UK, as of 25th August

Retail and Recreation: -11%
Supermarket and Pharmacy: -14%
Parks: +43%
Public Transport: -44%
Workplaces: -48%
Residential: +13%

Yes @Nellodee that would be both interesting and useful. They're at the sharp end and it would be so informative.

Eg I saw someone on MN say that her relative was advised by them to look out for symptoms including sickness, which I thought was significant as it's not one of the three "official" NHS symptoms! I'm also sending people the WHO or Zoe app symptom list.

Was referring to the "peripheral tracking data" you mentioned @Nellodee

Sorry to confuse with quickfire double posts!

[quote MRex]@sunseekin - DHSC used Ipsos MORI to get the random representative sample from the NHS database of everyone registered woods over the age of 5 to send out letters; this is info from the pillar 4 overview because I found that one first, but they were used with ONS too (not Imperial though, that one's UCL):
www.thelancet.com/journals/landia/article/PIIS2213-8587(20)30268-0/fulltext.

It's not like Ipsos Mori have ever got election nor referendum results wrong . They have all the usual demographic info to calculate what counts as representative and pick random individuals within that, so it's clearly been set up with the aim of not being biased towards any specific regions / age groups etc.[/quote]
Thanks for this, really interesting, p.s. am enjoying your over posting!

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