Daily numbers, graphs, analysis thread 13

[PatriciaHolm]

Taking the liberty of starting a new thread as we've just bust the old one, with much thanks to @BigChocfrenzy and I will copy her header..

Welcome to thread 13 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 📈📶👍

Median ages of population:

Developing countries have v low average ages, especially in Africa, where HIV has wreaked dreadful carnage

"Emerging economies" like Brazil have ages between those of developing and developed countries, also a more robust health service
At the very least, wilfully refusing all lesser measures, as Brazil's Trump has done, is utter abandonment of responsibility of the most basic duty towards their people

In Europe:
The UK has a lower median age than most other W / N European countries due to immigrant communities not yet having many very elderly
Ireland has an even younger population
So govt strategies, decision-making and public services are additional factors to consider when looking at IFR

Age Country
47 Italy
46 Germany
45 Spain
42 France
41 Sweden
40 UK
38 Ireland

32 Turkey
38 China

38 NZ
38 Australia

32 Argentina
33 Brazil

17 Uganda
18 Nigeria
22 Ghana
25 Egypt
28 S Africa
28 India

https://www.worldometers.info/world-population/population-by-country/

I don’t know how to reply to a certain comment sorry but someone further up has said there’s approx 1500 deaths a day in the Uk - are the coronavirus deaths now extra to this or should be considered part of that? As surely some of them would’ve died anyway of other causes? (Statistically?)

Very interesting twitter thread from the ONS here looking at excess deaths across Europe.
twitter.com/ons/status/1288754396977942528?s=21

@twolittleboysonetiredmum

I don’t know how to reply to a certain comment sorry but someone further up has said there’s approx 1500 deaths a day in the Uk - are the coronavirus deaths now extra to this or should be considered part of that? As surely some of them would’ve died anyway of other causes? (Statistically?)

...... The UK - and nearly all other European countries excl Russia - has been in the "Normal" range of deaths for several weeks

Some of the very elderly would have died shortly; some wouldn't
Only about 25% of the dead were in care homes with an average exoectancy of 2 more years and those outside would normally have much longer
There have been statistical analyses indicating several years of life lost per person

This is why governments - including the UK - and public health experts around the world have said that
"excess" deaths are the best measure of a pandemic
i.e. the number of deaths above the historic average of several years for a particular time period

The UK had an excess of 60,000 over a period of 3 months at the height of the epidemic
and like most other countries, came down to normal levels before lockdown was significantly relaxed

• which should end the myth that a significant net number of deaths so far resulted from lockdown itself We may well see thousands of extra deaths from recession over the next few years, but spread out over years they are not something that hammers the coutry as a whole - as with the earlier deaths from austerity and benefit cuts, it will fly under the radar for most people

@sunseekin

Someone kindly pointing me in the direction of this data on this thread...

coronavirus-staging.data.gov.uk/testing

Does anyone know if the daily number of tests processed (154496 for yesterday) corresponds to the positive coronavirus tests for that day.

Or if there is a time lag between the two numbers?

Large number of tests yesterday but smaller number of positive tests - hoping a positive sign for August but also wondering whether numbers match or not?

Thanks very much!

@sunseekin as far as confirmed cases go this number is simply that added to the cumulative total of confirmed cases, it bears no relation to the actual number of confirmed cases with a specimen date of 29th July - it will be several days before we get an accurate fix on what that might be

Taking yesterday's number of 763 cases added, 700 of these were in England, and related in terms of specimen date:

28th July 40 cases added
27th July 447 cases added
26th July 29 cases added
25th July 52 cases added
24th July 34 cases added
23rd July 45 cases added
22nd July 16 cases added
21st July 21 cases added
20th July 6 cases added

1st to 19th July 10 cases added

June - 5 cases removed
May - 6 cases added
Apr - changes net to zero
Mar - 1 case removed
Feb - no adjustments

Worth noting that the beta staging site will become the official site from 4th August.

coronavirus-staging.data.gov.uk/new-service

and that the beta staging name will be de-commissioned, so rather than being

coronavirus-staging.data.gov.uk/

as the new official site it will become

coronavirus.data.gov.uk

Also @sunseekin That 154,496 for yesterday incorporates all 4 types of test, not just the 2 that test for current infection (and thus are the tests producing the numbers of new infections announced). The total of Pillar 1 and Pillar 2 for yesterday was 130,611.

@PatriciaHolm are you aware of any more granular detail on testing? From the statement in "about the data" I'm assuming that the "daily" increase in the cumulative number tested is, like confirmed cases, an aggregation from a number of previous days?

Daily and cumulative numbers of tests
The number of tests, across all types of testing, reported on or up to the latest reporting date.

At least for confirmed cases there is a daily data file providing a complete history covering every LA allowing daily cases, by specimen date, for each one to be easily identified. I've not seen anything similar - ie for each LA - for test numbers by pillar - although I've not looked that hard, and am being rather lazy by posing the question on the thread :)

Here is another interesting twitter thread discussing testing and increase in cases which seems reasonable to me?
twitter.com/stevebrown2856/status/1288763251161997314?s=21

[quote boys3]@PatriciaHolm are you aware of any more granular detail on testing? From the statement in "about the data" I'm assuming that the "daily" increase in the cumulative number tested is, like confirmed cases, an aggregation from a number of previous days?

Daily and cumulative numbers of tests
The number of tests, across all types of testing, reported on or up to the latest reporting date.

At least for confirmed cases there is a daily data file providing a complete history covering every LA allowing daily cases, by specimen date, for each one to be easily identified. I've not seen anything similar - ie for each LA - for test numbers by pillar - although I've not looked that hard, and am being rather lazy by posing the question on the thread :)[/quote]
hmm interesting. The daily increase is, I think, as you say, simply the number that the total of reported tests increases by each day. It is not related to the day it is announced (as in, the 130,000 P1 and P2 tests announced yesterday don't relate to tests taken yesterday, or the day before; in fact, tests are counted differently depending on how they are delivered/processed (in person P2 test are counted on the day they are processed in the lab, at home P2 are counted when despatched)).

There is more detail here -
www.gov.uk/government/publications/coronavirus-covid-19-testing-data-methodology/covid-19-testing-data-methodology-note

Which makes trying to calculate any true level of positivity per day impossible. My previous slide is very simple and just looks at tests vs cases per announced day, on the assumption that it will show growth anyway, and seven day averages even more so.

In terms of granular location for tests, I've not seen any I don't think. I've just had a another look through the usual sources and can't see any.

[quote MarcelineMissouri]Here is another interesting twitter thread discussing testing and increase in cases which seems reasonable to me?
twitter.com/stevebrown2856/status/1288763251161997314?s=21[/quote]
Indeed - though I'm assuming he's normalised by overall increase in testing rates as, as just discussed, I don't think we have testing rates by area.

In reality the testing is likely to be quite localised by now, particularly the Pillar 2, so it's hard at local level to really unpick, but if that is the case then the picture would be better than he suggests as case positivity rates would be lower in the regions where we are testing more...

@PatriciaHolm - thanks.

I think the testing data by LA must be known, but is simply not published. I thought I had seen in one of the PHE weekly surveillance reports (noting the one due out today not yet published) the test numbers for the "concern" LAs, but may well be mistaken as nothing like that in the wk 30 report; just the list of "concern" LAs and direction of travel.

Thanks for the clarification BigChocFrenzy

[quote boys3]@PatriciaHolm - thanks.

I think the testing data by LA must be known, but is simply not published. I thought I had seen in one of the PHE weekly surveillance reports (noting the one due out today not yet published) the test numbers for the "concern" LAs, but may well be mistaken as nothing like that in the wk 30 report; just the list of "concern" LAs and direction of travel.[/quote]
I honestly don't know at this stage ;-) TBH there must be testing data by laboratory; and I would imagine there is data, somewhere, by drive/walkthrough testing centre as well. However, these will not necessarily line up with LAs, as people in one LA could easily get tested in another.

It must be statistically possible to measure testing at home postcode level, as you have to give you address for a test I think, but I wouldn't be surprised if no-one is bothering at local levels.

@PatriciaHolm aren't all the numbers driven by home postcode, so physical location of test, or indeed of any hospitalisation (not sure Rutland has any ICU beds :)) should not impact on ability to identify area of residence - providing of course the correct home postcode in provided / entered

[quote boys3]@PatriciaHolm aren't all the numbers driven by home postcode, so physical location of test, or indeed of any hospitalisation (not sure Rutland has any ICU beds :)) should not impact on ability to identify area of residence - providing of course the correct home postcode in provided / entered[/quote]
Well, we do know postcodes of everyone tested, so in a perfect world we could do analysis this way.

There are also de-dupe processes run regularly to remove duplicated results from NHS and PHE which require personal identifiers, so the data must exist altogether on one system somewhere. The question is whether anyone is bothering to try to sort it into LA, as this would be a significant task!

Anyone know why the NHS deaths in hospital numbers haven't been released yet today? They're normally pretty efficient at getting them out at 2pm, but there's no sign of them today...

@Shesingsshangrila

Anyone know why the NHS deaths in hospital numbers haven't been released yet today? They're normally pretty efficient at getting them out at 2pm, but there's no sign of them today...

They are out - 12 today.

846 positive tests today. That will take the 7 day average up as the equivalent last week was 769. It's also the highest one day since 29 June.

Tests look to be about the same as yesterday.

Staging site not updated yet so can't see where rises might have been and how the tests are spread over the past few days.

I think saturday July 15th was 888, it was an outlier

@hopefulhalf

I think saturday July 15th was 888, it was an outlier

coronavirus-staging.data.gov.uk/cases

Says 538 for Sat 15!

Ok my mistake. Worrying

Where do you see that Patricia ?

Not seeing today's figures

www.gov.uk/guidance/coronavirus-covid-19-information-for-the-public

Stats seem to appear here first, but only top level cases and tests.