Daily numbers, graphs, analysis thread 9

[Barracker]

Welcome to thread 9 of the daily updates.

Resource links:
Worldometer UK page
Financial Times Daily updates and graphs
HSJ Coronavirus updates
Johns Hopkins Coronavirus Resource Centre
NHS England stats, including breakdown by Hospital Trust
Covidly.com to filter graphs using selected data filters
ONS statistics for CV related deaths outside hospitals, released weekly each Tuesday

Thank you to all contributors for their factual, data driven, and civil discussions.

Last Tuesday's figure was 136 (but after bank holiday so way out) and the previous Tuesday was 545 ... so yes, a heck of a drop

BigChoc
I take the point that this thread is about the UK
but posters regularly compare with data from other countries
who may not have the same methodologies
hence my concern

The UK stats agency letter was at least the second one, I believe. The first (or more) was/ were not made public but there was reference somewhere to a reply.

Listening Noone compares the UK to Tanzania, which is the country you mentioned.

It is almost always to other European countries and we do say who includes care homes and who - Belgium - include any suspicion of COVID anywhere

The comparisons aren't exact, but give a reasonable idea

"the previous Tuesday was 545"

Puzzled On that Tuesday, I think PHE added a large catchup of 346 deaths,
in which case it's not suitable for comparison

true figure was then about 200

Ah - fair enough, BigChoc; even the death figures are leaping around so much now that it's hard to keep track

@BigChocFrenzy

"the previous Tuesday was 545"

Puzzled On that Tuesday, I think PHE added a large catchup of 346 deaths,
in which case it's not suitable for comparison

true figure was then about 200

I don't think that is quite right Bigchoc, that 346 was "just" the number of deaths that occurred outside of hospitals reported on the 19th. Today's figures have a similar addition of 172.

Friday it was 151, yesterday was a negative number (as some deaths get counted in community and hospitals)

Some arsehole shit-stirring journalist claimed that a covid-19 report had been censored because it would inflame racial tensions

twitter.com/SamCoatesSky/status/1267568553177288710

In fact it turns out to be 89 pages of graphs and references and doesn't make any assertions and has no chance of inflaming any tensions whatsoever.

It's also not specifically about BAME, and mostly in fact is not. It covers age, care homes, BMI, region, deprivation, etc.

It is worth reading in full

www.gov.uk/government/publications/covid-19-review-of-disparities-in-risks-and-outcomes

Main point is it does show that black people are most likely to have been infected with covid-19. But it does answer a lot other questions as well, albeit that those related to deaths have all been answered in this thread already. I'm not sure the infection data by race is new though.

I.e. I think there may be much new data about infection demographics in the report, although this might just be because I tend to ignore those data as unreliable as having massive sampling bias. There's a lot of old material syntheised also however

Littlebelina I think the 346 deaths were catchup,
because that 545 looks an odd spike in the graph after 2 low death days of well under 200 :

Yes but it's the usual Tuesday catch up, you always get low numbers on a sunday/Monday and higher on a Tuesday as not as many deaths get reported on weekends (offices closed etc). It's worse for the non hospital figures (those 346), presumably as registry offices are closed. It's why we should look at rolling 7 days or only compare a day with the same day a previous week (which is what puzzles was doing). Last week the catch up was Wednesday due to bank holiday so she went back two weeks to compare Tuesday with tuesday

That PHE report on risk disparity is what I posted upthread, but I saw nothing particularly new wrt race

• and I have a personal interest in assessing that risk.

I posted mainly the 70 x risk they estimated for age 80+ to under 40,
not because it's new but because the magnitude still shocks me
(even if we didn't calculate exactly 70 x before, we knew the huge risk factor)

The same thing is evident the week before. Yesterday we did have a spike due to a historic adjustment (evident in the graph). The difference was that this was reported separately, the figure for the day was 111 but an additional 400+ were added to the total (and the person drawing this graph added it to yesterday's total so the weekend effect isn't as sharp

I wouldn't compare catchup days; they vary too much

We should compare the 7-day rolling average, or look at the overall curve form

Puzzled not puzzles sorry

The data is too noisy to place significance on individual days

In Germany, total cases remain low and not yet a matter of concern

However, the last few days have illustrated again the higher risk of certain activities with a lot of people in proximity

• care homes, meat plants, churches, parties and a kindergarten - and there have been small localised lockdowns

https://www.rki.de/DE/Content/InfAZ/N/NeuartigesCoronavirus/Situationsberichte/2020-06-02-de.pdf??__blob=publicationFile

"In the city of Bremerhaven, an outbreak occurred after a Pentecost service.

In the district Sonneberg, an outbreak occurred in a nursing home for the elderly in which several residents and employees were infected with SARS-Cov-2.

There was also one case each among employees of an outpatient nursing service and a kindergarten.
In Coburg outbreaks in long-term-care-facilities were reported.

A Covid-19 outbreak occurred in the district of Göttingen, probably due to transmissions at private family celebrations
and for which a large number of contact persons has been identified.

Outbreaks continue to occur in meat-processing plants in several federal states,
some of which have led to production closures.

Further outbreaks in religious communities are reported from Berlin and from Hesse."

Hope you don’t mind me posting a question here but thought it might be the best place to ask. Sorry if it’s stupid one but can’t find the answer on google.

When looking at the number of cases in my area it has crept up over time but not by much over the past few weeks. Does the cases per area shown by the BBC, Telegraph etc. mean they are all active cases? Surely the number of active cases must be lower than published and a lot will have had an outcome and are no longer an infection risk?

Thanks

Meat processing plant outbreaks are significant because it suggests that the virus spreads better in cold conditions and could therefore be seasonal.

I am beginning to get frustrated with the media headlining BAME deaths or schools, and keep wanting to shout the real story is in age. (70x more likely to die if over 80 fgs!).

The question the media should be asking the government is what is their strategy to protect the health and preserve the freedom of those in the 65+ age range.

Even the media coverage on nhs workers deaths masks that most deaths are coming from the retired population.

FT has updated its chart of excess deaths in several countries

The UK had 62,000 up to 22 May

Peru is the 2nd worst hit country in S America after Brazil,
but its health system is also struggling to cope with an epidemic of Dengue fever at the same time
Hence high excess deaths in Peru

howdidwgetthere can you post a link to the data you're referring to?

www.google.co.uk/amp/s/www.bbc.co.uk/news/amp/uk-51768274

@EugeniaGrace a very valid point on age.
What do you expect the government to do to preserve the freedom of the 65+ age group?
Surely the answer is what they are supposedly trying to do - social distancing, get the overall infection rate down, find a treatment and as fast as possible a vaccine.
Do you have anything specific in mind?

howdidwegetthere yes that's a cumulative number by your Upper Tier Local Authority, so that's how many in total confirmed cases (ie tested) - likely to be more out there untested - but including all from Feb onwards until now, so many will have recovered/died.

Your best bet if you want to see cases over time is to download the actual data here coronavirus.data.gov.uk/#category=utlas&map=rate under the map there's a link to download cases or deaths data by CSV. I've been using these to make my own charts for local areas.