Daily numbers, graphs, analysis thread 12

[BigChocFrenzy]

Welcome to thread 12 of the daily updates

Resource links:

Slides & data UK govt pressers
UK dashboard sub-national data, local authorities
Beta Uk dashboard deaths, cases, hospitals, tests, partially sub-national
UK stats updated daily by PHE & DHSC
ONS UK statistics for CV related deaths, released weekly each Tuesday
PHE surveillance report infections & deaths released every Thursday with sep. infographic
NHS England stats including breakdown by Hospital Trust
FT Daily updates
HSJ Healthcare updates
Worldometer UK page
Plot FT graphs compare countries deaths, cases / million pop. / log / linear
Covidly.com filter graphs compare countries
Plot COVID Graphs Our World in Data

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

Thank you for this. These threads have kept me sane.

Thanks, BIgChoc

Thanks

Thanks BigChoc

Thanks @BigChocFrenzy

Thank you.

Thanks for the new thread.

Thank you!

Checking in

Has anyone looked closely at the cases by nation data on the Gov beta site?
coronavirus-staging.data.gov.uk/cases?areaType=overview&areaName=United%20Kingdom

Since they reset the cases and discounted all the double tests etc it shows that the daily cases were very low in the last half of June around 200-250. The more recent results, which I'd thought were dropping reassuringly, actually appear much higher in comparison (although still low compared to the 1k plus numbers we were used to hearing.)

Looking at the local authorities, Leicester in particular makes up a significant proportion of the cases. So extra testing there, Kirklees and Wrexham may contribute to that.

Thanks

Where did you get that chat from exactly @Cathy? The one I can see on the Beta site (not broken down by country) looks quite different. It doesn't show such low rates in early June, nor the recent bulge like in your picture

@cathyandclare

I’m a bit confused by that graph and double-counting. Even if all tests had been double-counted, we were regularly reporting 1,000+ test cases in late June so I’m not sure how the figures become 200-250?

@cathy - I've now seen where that graph is. It doesn't make sense though - The one just above it on the same page in the Beta site not broken down by country looks quite different. It doesn't show such low rates in early June, nor the recent bulge like in your picture. They both have the same title (cases by date reported) so they should match (unless there's some huge number of cases in some 5th UK nation!) Anyone know why they don't match?

@Firefliess it's on the page linked: coronavirus-staging.data.gov.uk/cases?areaType=overview&areaName=United%20Kingdom

It's titled per nation. Thinking about it, I wonder if it was just the Pillar 1 data before the data counting reset and now it covers Pillar 2 as well, hence the increase.

Delving into the 'about the data' section on that beta site it says this about daily and cumulative cases:

The UK total is not the sum of the 4 National totals as the pillar 2 cases are not included in the National totals for all parts of the UK - they are currently included for Wales, but not England, Northern Ireland or Scotland

I thought the Pillar 2 testing was included for England, it has been all along hasn't it? We just didn't know the regional breakdowns.

@cathy - I think you might be right that it's pillar 2 only until the last few days - because we knew they were having trouble breaking down the pillar 1 tests by country. It's a crazy chart to publish though, without explanation. Yes I know they're under pressure, but really? Two charts on the same page both showing daily cases and with the same title showing completely different data?

Marking place. Thank you.

Two charts on the same page both showing daily cases and with the same title showing completely different data?

This pandemic has been a fascinating case study into the difficulty of creating, updating, and releasing accurate and timely data.

The conundrum of how to count tests of people when some people are tested regularly and others rarely, if at all, is fascinating. It doesn't seem complicated, but it is.

@Firefliess

I think you might be right that it's pillar 2 only until the last few days - because we knew they were having trouble breaking down the pillar 1 tests by country.

I think you’re right, but i think you have pillars 1 and 2 the wrong way round.

Daily cases by specimen date, and cases by Nation - appears to be Pillar 1 and 2, for each date (you can tell by cumulative total). This is actually more interesting than by date reported anyway, as it's more relevant to have actual day of test. Last couple of days will suffer from data lag though.

Daily cases by date reported -appears to be just Pillar 1 until July 2.

@PatriciaHolm

Daily cases by specimen date, and cases by Nation - appears to be Pillar 1 and 2, for each date (you can tell by cumulative total). This is actually more interesting than by date reported anyway, as it's more relevant to have actual day of test. Last couple of days will suffer from data lag though.

Daily cases by date reported -appears to be just Pillar 1 until July 2.

arghhh sorry by Nation is also only Pillar 1 until July 1!!

Spiegelhalter: Risks, R numbers and raw data: how to interpret coronavirus statistics

Useful perspective on stats - and being v careful of media headlines

https://www.theguardian.com/world/2020/jul/05/risks-r-numbers-and-raw-data-how-to-interpret-coronavirus-statistics

He has a good diagram of tests & "true status" which he tweeted to accompany the article

"Pubs close again after punters test positive for COVID-19"

In case anyone is worried about this story, it is a TINY number compared to the umpteen thousand pubs that reopened

The pubs behaved responsibly, contacted punters

• which illustrates the importance of having contact details -

and closed for deep cleaning, staff testing.

That is how to squash v minor outbreaks after lockdown ends

https://news.sky.com/story/coronavirus-pubs-close-again-after-punters-test-positive-for-covid-19-12022747

ONS figures are out for the week ending June 26th:

The number of deaths registered in England and Wales in the week ending 26 June 2020 (Week 26) was 8,979, this was 360 deaths lower than Week 25.

In Week 26, the number of deaths registered was 3.4% below the five-year average (314 deaths fewer), this is the second consecutive week that deaths have been below the five-year average; the numbers of deaths in care homes and hospitals were also fewer than the five-year average (103 and 815 deaths lower respectively), while the number of deaths in private homes was 745 higher than the five-year average.