Daily numbers, graphs, analysis thread 8

[Barracker]

Welcome to thread 8 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.

The mutliple seed points look to be "community spread" rather than direct infections from abroad
Wide community spread is a sign that the infection has been allowed to run too long without effective measures being taken.

Big choc our particular issue was large parties of (young healthy) people going skiing at half term and coming home all within a few days of each other. I guess many of those people were asymptomatic and began community spreading. It seems like there was already undetected spread beginning in London but I think this exacerbated the way infections in many other towns and cities all moved forward at similar times - rather than spreading out from one area as happened in Italy, Spain etc...

It makes sense that multiple "seats" of infection plus late lockdown = perfect storm...

There were ski trips in several other countries, especially Germany I know

It's not a particular difference wrt the UK vs other European countries, many of whom have rather more skiers.

I'd never been aware of so many Italian ski trips as I was then!

I bring you a highly suspicious fact checker doing the rounds. I'm not brainy enough to even begin to pick through it; a few claims I know to be wrong from
here and obviously statistics and in the eye of the beholder. (Two peeps I know criticised it! But I don't know much more.)
There's some concern as to who is finding it.

https://swprs.org/a-swiss-doctor-on-covid-19/?fbclid=IwAR0NlgbWQNv2zF17sQdWi6jvFWo78P0UrRFikJELL_HV1YcYSPh7PM12kC4

I remember the early threads here and that led to this one; locking down quickly was being discussed a lot. Dh (obvs no one important but like a stuck record) has said all the way through we were too late.

This is worrying - the first evidence of an official attempt to doctor COVID data - and it's in the USA:

Architect of Florida’s coronavirus dashboard says she was fired for not doctoring data

Our main defence against government malfeasance is probably whistleblowers

https://nypost.com/2020/05/19/coronavirus-dashboard-architect-claims-she-was-fired-for-not-changing-data/

In an email to CBS12 News, Jones said her removal was “not voluntary” and that it was
because she refused to “manually change data to drum up support for the plan to reopen.”

She announced her ouster last week in an email to colleagues, saying her department is no longer responsible for updating information on the site “in any shape or form.”

@itsgettingweird Interesting briefing today. Anyone get the feeling that scientists are kicking back against being blamed for every decision? Very much so, can't remember the scientist's name, Dame Angela something? She was definitely kicking back and it wasn't very subtle was it. At the end I thought, "I bet she won't be on again." Mind you after Therese Coffey this morning I don't blame her.

She's Deputy Chief Scientific Adviser Professor Dame Angela McLean

Scientists are paid to advise, not to make these national decisions - the government is elected to make those

until the last couple of pages this has been a well mannered thread it would be good to keep it that way. People are entitled to put forward their own speculative comments on what has been happening and for replies to stay polite. We'll only be able to learn the full lessons from this a lot later when it's possible to dig deeper into the data.

I have no idea what other countries do at half term but it's pretty clear that not only did a lot of young adults go skiing fro Britain but a lot of school parties did so too. There are probably a lot of British children that have had an asymptomatic dose of this virus and given it to their younger siblings and school friends. Some secondary schools could probably go back now with very little impact on their teachers or their local area.

Football matches, Cheltenham races, church series and religious festivals all played their part here - and religious services certainly had an impact elsewhere.

That Swiss doctor certainly has an agenda. I read some questionable parts and decided it wasnt worth my time.

BigChocFrenzy
There's about as much evidence as for thousands of people in each of those countries dying in previous years,to make the UK performance look less disastrous.

I think you are deliberately misunderstanding me - I was querying comparisons of excess deaths with France, Spain, Italy...who also have high numbers of COVID-19 deaths. I have already posted; "Obviously this still doesn't reduce the number of excess deaths in the UK..."

What were the actual deaths (not estimated excess deaths) per day/week for France (similar population size to the UK) in March/April/May 2015-2019? Do you have a link?

PPE
Even Germany reported 11,963 cases of COVID infection among staff working in hospitals and outpatients facilities.The UK's lack of PPE compared to other European countries has been exaggerated by the media.

@BigChocFrenzy Thanks, I couldn't think of her surname. I do like her, I think she is my favourite of the Scientific Advisers.

There were thousands of skiers from Germany too, schools and young adults.
hence all the reports that the infection age was unusually young here in the early part of the epidemic, compared to e.g. the UK

Many international trade fairs in Germany too, as well as international sports events.

re possible causes for the very high UK death rate:
we need to stay with facts, rather than unsubstantiated fantasy

Speculation about mass spring deaths in previous years on the continent

• that would also render the government's own previous comparisons invalid - does need to be backed up by data It would be easy enough to find some, if any such events actually happened, so let's park the issue until then.

International papers I've read - USA, India, Malaysia, European - have run several articles blaming poor / late UK government decisions and disorganisation,
but imo an inadequate public health system and associated infrastructure played a big part too.

The most basic lesson in business is that until you accept there is a problem, you can't solve it.

A question for the thread experts
and my apologies if its already been linked loads
Is there a time series graphs of deaths per month covering several years

• so that its possible to see the seasonal fluctuations

• and the impact of past bouts of infectious disease on different age groups ?

If not, is the core data set easily available to chuck into a graph ?

The weekly death stats are published by year here

www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/datasets/weeklyprovisionalfiguresondeathsregisteredinenglandandwales

There are 10 years worth there but you'd have to compile yourself

Weekly fluctuations can be misleading due to bank holidays making registrations get.pushed back from certain weeks over Christmas, Easter etc

"PPE
Even Germany reported 11,963 cases of COVID infection among staff working in hospitals and outpatients facilities."

As I've repeatedly replied to this, German health staff have been intensively tested from early on.
Hence found almost all the cases, instead of the usual 1/10, 1/20 or whatever in the rest of the country

The low number of deaths of medical staff, only 18, indicates that health staff were at lower risk than the general population of working age.

It is more likely that they were infected at the supermarket etc than in the enclosed space suits that they've been wearing for COVID patients

"What were the actual deaths (not estimated excess deaths) per day/week for France (similar population size to the UK) in March/April/May 2015-2019? "

No, but you could Google to find their ONS equiv
If it is your theory that there were a large number of Spring deaths in say France for those years, then do dig up the data

Shootsfruit
I started to build a query on here
www.nomisweb.co.uk/query/asv2map.aspx
I reckon it could be done to track the impact
because calendar years do not match seasons ....

Ah yes...the usual squirrels here 👀

What were the actual deaths (not estimated excess deaths) per day/week for France (similar population size to the UK) in March/April/May 2015-2019? Do you have a link?

This doesn't give the data you are asking for, but does use the expected deaths and excess deaths to make a p score (excess deaths divided by expected deaths). This utilises the figure you are looking for - expected deaths (the average of the previous 5 years) to make a figure which shows excess deaths as a proportion of expected. They also do something similar to look at whether deaths in a particular country are happening in the age groups you would expect in the numbers you would expect.

You could look at Euromomo too.

TBH I'd not be interested in any time series with groupings of less than a month because of the 'noise' issue raised by shootsfruit
but the fluctuations of deaths per 100,000 per 10 year age group over time
would probably highlight excess deaths rather effectively
and be comparable across time and between countries

I'm just not nearly as good at the database queries as others !

Listening It would be a superb project to do, suitable maybe to while away your lockdown ? < hopeful >

BigChoc
I'm not locked down - I'm running around like a mad thing working -
but yes, when life calms down a bit I might download the data sets and have a play or I might set one of my children on it as they are better at databases than I !

France death data

If you dig around on this site - the French equiv of ONS - you should find the raw data on deaths for previous years:

https://www.insee.fr/en/statistiques/2555227?sommaire=2382613

re possible causes for the very high UK death rate:
we need to stay with facts, rather than unsubstantiated fantasy

You are still misunderstanding me - I am querying the comparison of excess deaths with other countries at this particular time of year.

Speculation about mass spring deaths in previous years on the continent

It depends what you mean by 'mass spring deaths' - the ONS observed 243 more deaths in the UK on two hot days in April 2018. How does that compare to Italy or Spain in an average year? Just 100 extra deaths per day becomes 7,000 deaths in 10 weeks. I also linked to evidence reporting the UK's influenza season typically peaks in January instead of February and March - therefore you might expect fewer March deaths from influenza/respiratory diseases in the UK compared to some other European countries and their average death rate might be higher at this time of year. Equally, you might expect more people to die from summer heat in the South of France, Spain and Italy compared to the UK, but not necessarily more people to die in those countries over the whole year compared to the UK.

It is more likely that they were infected at the supermarket etc than in the enclosed space suits that they've been wearing for COVID patients

"German doctors pose naked in protest at PPE shortages."

www.theguardian.com/world/2020/apr/27/german-doctors-pose-naked-in-protest-at-ppe-shortages