Daily numbers, graphs, analysis thread 4

[Barracker]

Welcome to thread 4 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.

What I've read is that countries need about 95% compliance for lockdown for it to work properly,
or it has to be extended / tightened / released before CV controlled

The lack of precision in the UK rules may be political compromise rather than scientific planning

There seems to be a 50% govt target floating around MN,
but this is not based on anything I can find, other than wishful thinking

https://www.theguardian.com/society/2020/apr/10/uk-coronavirus-deaths-bame-doctors-bma

The head of the British Medical Association has called on the government to urgently investigate if and why black, Asian and minority ethnic people are more vulnerable to Covid-19,

after the first 10 doctors in the UK named as having died from the virus were all BAME.

Those doctors have ancestry in regions including Asia, the Middle East and Africa

Even allowing for the overrepresentation of BAME staff in the NHS
– they comprise 44% of medical staff compared with 14% of the population of England and Wales –

the fact that they were all from ethnic minorities was “extremely disturbing and worrying”
.....
“Not only that, we also know that in terms of the BAME population, they make up about a third of those in intensive care.

There’s a disproportionate percentage of BAME people getting ill.

www.icnarc.org/DataServices/Attachments/Download/41bbc4bb-2c7b-ea11-9124-00505601089b

3rd ICNARC report is out.

Now looking at a larger data set, but the survival rate of critical CV19 patients is still more or less 50/50.

Again, these figures are based on shorter than average stays in the ICU -we should begin to see some data based on longer stays next week,, I would imagine (the 4th weekly report)?

The additional deaths could either be undetected Covid deaths or other deaths occupying because of Covid eg people not going to hospital with strokes of heart attacks.

There tend to be more deaths when seasons change in spring and to a lesser extent in autumn. It’s colloquially known as stroke season in healthcare, so regardless of Covid 19 a change of temperature would usually bring with it a rise in deaths.

I was in the Range 2 days ago. There were no restrictions or requirements to buy particular types of items. Lots of people were buying things like plants, compost, gravel etc. Some were browsing the home decor isles ( but not very many). I saw a lady leaving with a trolley full of garden toys as I arrived.
They do sell toiletries and cleaning products which are essentials but they didn't have any tissue which I wanted

Our local police are apparently staking out (I think that's the phrase) in the Range car park, checking for non essential purchases! It's in the local online paper.

I'm sorry if this has already been posted:

www.ons.gov.uk/file?uri=%2fpeoplepopulationandcommunity%2fbirthsdeathsandmarriages%2fdeaths%2fdatasets%2fweeklyprovisionalfiguresondeathsregisteredinenglandandwales%2f2020/publishedweek132020.xlsx

Please take the time to look at this carefully and remember that dying with coronavirus and dying from are two very different things.

Ireland

https://www.irishexaminer.com/breakingnews/ireland/nursing-home-residents-make-up-more-than-half-of-irelands-covid-19-toll-993501.html

More than half the victims of Covid-19 in Ireland were nursing home residents, the chief medical officer has said.

Of the 288 people who have died so far in the outbreak, 156 had been living in nursing homes prior to their deaths (54%),
Dr Tony Holohan said.

Most of those deaths occurred in hospital settings
< so probably included in Ireland's total >
......
Dr Holohan said in the affected nursing homes,
one in five staff members had reported Covid-19 symptoms,
and one in six residents.

"dying with coronavirus and dying from"

Many Western countries are using dying "with" coronavirus
because unless they were terminally ill, coronavirus caused premature death

Much like if you shoot a terminally ill person, the cause of death is murder by shooting

Regarding compliance, here are the assumptions Imperial used to create their model. They assumed social distancing would lead to a 75% reduction in community contacts, and only a 25% reduction in workplace contacts. I think these percentage reductions are much greater in reality (if you think how many people you crossed paths with in a typical pre-corona week vs now!). They assumed only 50% households would comply with the 2-week quarantine of a symptomatic household and only 75% of over 70s would comply with the self-isolation. Again, it seems to me that the UK have been much more compliant than these assumptions.

Biggest wrong assumption - the one that is really worrying them - is that far more people than they expected stopped work if they could not WFH

That both buggered up the economy more and cost the Exchequer far more in benefits / 80% pay

So the lockdown is costing billions more per month than they planned

Interesting that the Daily Fail graphs for show a pattern. Nationally: for two days numbers fall then theres a one day increase - & then it repeats. London: two days in a row numbers go up then a one day drop- & then it repeats.

The Imperial model estimated that lockdown would reduce the death toll to about 20,000

So if the deaths are around that, then the lockdown was about right

If it's much more than that, then the model was too inaccurate / the lockdown was not strict enough / more people are being non-compliant than MNers realise

If it's much less .... then each life saved cost more than they expected / were prepared to pay

Am I right in thinking that the trend is putting the deaths way way over 20k unless it drops off very quickly?

I also think, 2/3 weeks after Easter weekend the figures will rise sharply again.

John Burn-Murdoch@jburnmurdoch (FT stats geek)

Fri 10 April update of coronavirus trajectories

Daily new deaths:
• US & UK still on clear trend of rising daily deaths.
Reporting patterns may give occasional dips, but clearly trending up
....
Now cumulative deaths:
• US death toll still ramping up, and likely to become the highest worldwide if not tonight then certainly tomorrow
• US may end up off the chart again (above 500,000 cases) by the end of the night
. Turkey is battling one of the world’s most severe outbreaks
...
Now subnational region daily deaths:
• NY death toll still rising every day
• London the same
• Few places have had daily death tolls still rising this far in;
these two are now the two urban epicentres, globally
....
Subnational death tolls cumulatively:
• NY likely to have world’s highest subnational death toll within days
....
• We’re now showing all UK countries (Northern Ireland added);
England still accelerating much more steeply than the rest
....
Small multiples for daily new deaths in 46 countries:
• Norway locked down while Sweden didn’t; Norway’s daily death toll rising much more slowly than Sweden’s

Barracker He also posts:

"I’ve also changed data sources.

I’m no longer happy using Worldometers, so tonight I dropped them as even a secondary source and got data directly from official sources in all cases.

This takes longer, but given my concerns over data quality on covid, I think this is important."

Won't it be Wednesday or even Thursday that we get a set of figures that be comparable to today's? 2 bank holidays, I think social services ( so plus others) get Tuesday too. Some universities are getting Wednesday as well.
Doesn't bode well even given the inaccuracies we already have

monkeyToesOfDoom - I love your name !
It was suggested earlier on the thread that the UK might end up around 35,000 deaths

BUT
That is just until summer, when cases & deaths are expected to be v low throughout Europe

Noone has estimated the final toll in any country before - hopefully - there is a vaccine

BigChocFrenzy I also switched data sources a few days ago to the DHSC. That's why I'm only posting the update later in the day, because they release figures late.

To be fair to worldometers, they tend to go with the days release from the 4 nations which is out at 2pm. However the DHSC number never tallies with this for inexplicable 'timing' reasons. In fact I haven't yet seen a breakdown of how they arrive at such a different number from the total released at 2pm.

Lewis Goodalll@lewis*_goodall

NEW: Italian PM Guiussepe Conte says that the lockdown there will be extended until May 3rd.

After that he says, if the virus continues to lessen, life could gradually return to normal.

Just catching up, thanks for the new thread.

monkeyToesOfDoom - I love your name !
It was suggested earlier on the thread that the UK might end up around 35,000 deaths

Well glad you like my name :)

If death toll does end at 35k that will be horrific enough. But much lower than some estimates.

Has anyone done a comparison with Italy recently? Was it FATE doing them before?

We seem to not be tracking Italy so much anymore, which is why I stopped posting my data. I'm still tracking though. Here is a compare aligned from the date each country hit 50 deaths.

Italian PM has said a lot of industry/businesses can go back to work from 14/4 onwards. As predicted almost anything to kick off the economy.

Belgium is adding care home deaths into the total.
“40% of the deaths reported so far concern nursing homes or residential care centres and it concerns, mostly, people who did not have access to a lab test but who are treated as suspected coronavirus deaths,” Emmanuel André of the government’s coronavirus advisory team said in the press conference.