Daily numbers, graphs, analysis thread 6

[Barracker]

Welcome to thread 6 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.

Thanks shoot
That's a really informative anylysis and graphs

ICU

I've been trying to find out ICU outcomes for different age groups,
but the only real breakdown on that I've found so far is from NL

The length of time spent in ICU is surprisingly high, average 23 days, some still there at 40 days.
The very long stays show a smaller % then dying - maybe longterm stabilised, but on ventilators ?

So although the risk of actual death is v small for the young & middle-aged,
we should also consider the risk of a long period of illness and recovery

Charts:

1. Outcomes / length of stay
2. Outcomes / age

"How deadly is Covid-19 compared to seasonal flu, past pandemics, or car crashes?"
"It's about the spike"

https://www.thenewatlantis.com/publications/not-like-the-flu-not-like-car-crashes-not-like

Also compares to 1957-1958 Asian flu which killed over 1 million worldwide, 70-116,000 in the USA

1. deaths for the whole USA
2. for New York, the state hardest hit.

Note that the data sets begin at different points in the year (as marked on the left).
Also note that the figures shown here are for new deaths each week, not for cumulative deaths.

The charts have been updated on 21 April, since I first bookmarked the article
and COVID has since spiked to leading cause of US death atm

@Mustbetimeforachange thanks for link to commentary on modelling

CDC: Hospitalization Rates and Characteristics of Patients Hospitalized with Laboratory-Confirmed Coronavirus Disease 2019 — COVID-NET, 14 States, March 1–30, 2020

Gives hospitalisation rates for age, ethnicity
Note: CDC here recommends everyone wear "face coverings in public settings"

https://www.cdc.gov/mmwr/volumes/69/wr/mm6915e3.htm?scid=mm6915e33_w

Global coronavirus death toll could be 60% higher than reported

Mortality statistics show 122,000 deaths in excess of normal levels across 14 countries analysed by the FT

www.ft.com/content/6bd88b7d-3386-4543-b2e9-0d5c6fac846c?shareType=nongift

Reducing the lockdown. Is he correct? www.google.com/amp/s/news.sky.com/story/amp/coronavirus-over-100-000-could-die-if-only-vulnerable-groups-told-to-isolate-says-professor-11979126

Is that estimate for vulnerable (shielding plus elderly or what?) people continuing with lockdown rules & absolutely everyone else going back to normal? I didn’t think that was an option ... I assumed it would be vulnerable shield & everyone continues with social distancing +/- additional restrictions if/when required?

(I think that’s what he meant: Warning that no country has successfully shielded those most at risk from the virus while allowing the most resilient to continue as normal, he suggested that some kind of social distancing would need to remain in place until a vaccine is released. )

I hope so because that would be far easier to pull off than everyone on lockdown until a vaccine comes out of the wide blue yonder!

As of today there are only 43 COVID patients in ICU in the whole of Australia. Their lockdown seems to be working

I created this map showing extra deaths of the last 4 weeks relative to average deaths over 4 weeks.

If we consider only the last week, then the death total in any area might be low/high because of random noise, so caution is needed. But we should get a better picture of peak (or near peak) deaths.

Third image shows close-up on SW showing relative death total (colour coded relative to average weekly deaths) and actual excess death totals (for week 15). Larger numbers are more likely to be statistically significant than smaller - 12 excess deaths in an area with 100,000 people is less relevant than 120 excess deaths in an area with million people

Further close-ups here

imgur.com/a/JlWEFbb

Obviously where the latest week and the 4-week map shows no increase, it's likely that there won't be significant excess mortality in that area in the next few weeks, due to the lockdown.

Thanks for the contour plots and for the closeups again, shoots
Much appreciated

Even with the limited testing the UK is doing, the high positive % means

the UK is likely to overtake Germany in number of cases within a couple of days
and France by about the next weekend

to be #4 in cases in the world,
after the USA, Spain, Italy

The Uk is already #4 in hospital deaths (France has about 14,200)

There is a long gradual downward curve in deaths for all countries,
unlike the sharp gradient when deaths were increasing

I don't know if it would be feasible, politically and economically, to keep lockdown in UK areas with high infections & deaths,

but gradually release elsewhere

• Shoots contour maps indicate areas where has been little infection and little increase

It would probably also mean sealing off these red areas from outward travel
... which if spotted in advance would likely see those with 2nd home-owners buggering off, as in Italy, taking infection with them,
as well as causing great resentment

Probably only feasible in countries with a very large geographical area, such as the USA or Canada

Bigchoc,

I think it would be possible to lift lockdowns differently in different geographical areas in the U.K.

You would need to put something in place, even something light such as France, where you have to have a filled in form with ‘purpose of journey’. And there would need to be some enforcement with actual, real fines...

However, the differences would mainly be things like schools and work places opening. You can’t travel to take advantage of those.

In addition, most of the rich Londoners I know have already fled to their ( or a friend’s) foreign or country house. There would not be many left to flee.

I think that there is a bit of an obsession with inter country competition in death rates and also with proving the total death rate is higher.

It is almost impossible to compare countries due to a variety of confounding factors including population density and age of population. In addition, if the normal infectivity rate is 2.5, locking down 5 days later approximately quadruples the number of deaths.

I do think that the World, to date, has done remarkably well with only 200,000 deaths. Even if you double this (due to underreporting) 400,000 deaths is hardly noticeable in a population of 7,000,000,000 people. If we have passed the peak, the total deaths of the first wave should be well under 1 million, and probably under 750,000.

The question is where we go from here....

The decreasing gradient is going to be far lower than the increasing gradient. For the logarithmic gradient to be the same, you would need an infectivity rate of around 0.4; it is estimated to be between 0.6 and 0.9.

Interestingly, and positively, I wonder there will be a higher immunity rate amongst those most likely to become infected (health workers, mps, warehouse workers etc). That might mean that we may retain a lower R0 after unlocking (deconfining?). It does not only matter the overall proportion of the community who has immunity, it also matters who they are..

You could probably relax restrictions for most of Wales, I'm guessing Scotland as well, Devon, Lincolnshire, etc.

Then keep an eye on R0

Clearly places with a very high base like London must wait longer

Deaths from people living in Cumbria are probably being recorded against Newcastle , guess this is true for a lot of rural regions

The country boundaries do not make natural boundaries for peoples normal working lives

So people in south Northumberland often work in Newcastle

There are maps on the ONS based on something like 90% of the population of this region works in this region , these tend to be much bigger areas and almost certainly captures the hospital issue , might be more practical than a country level?

You'd want a strict ban on holidays in the UK for people from London, Birmingham, Manchester, Liverpool, etc.

"Deaths from people living in Cumbria are probably being recorded against Newcastle , guess this is true for a lot of rural regions"

No.

This has been discussed a few times in this thread. They are recorded against place of residence. The map is accurate in terms of place of residence.

People living in Cumbria who work in Newcastle will tend to lead different lives from those who live in Newcastle and work in Newcastle. Same reason that people who work in London but live in Guildford (which will be much of the population of Guildford) have much lower rates than Londoners.

I.e. commuters will be working office jobs. And quickly furloughed or working from home. Less socialising, etc.

It would probably also mean sealing off these red areas from outward travel.

I really don't see how you could seal off the red areas. Salford is a red area but the rest of Manchester isn't. How on earth would that work? I'm just imagining all us poor Salfordians lining the boundary roads sniffing the free air like modern Bisto kids.
If you open up workplaces in other areas of Manchester then Salfordians will need to be able to travel to them in order to go to work. Many children attend schools across the boundary, etc etc

You don't need to seal off areas!

The idea is to minimise transmission, not eliminate it.

Realistically, you could open schools and businesses at different times, not restaurants and places of entertainment. You cannot travel to a different area for school or work.

If there were sealing off areas, it would probably be sealing off roads to quite large regions,
such as entire cities, towns, counties

I can't imagine "COVID walls" within GB cities, like the "peace walls" in NI