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.

Here's a chart showing death rates by London borough (adjusted for population profile relative to total deaths)

Next chart shows the trend between white British % and relative death rate.

Note that the two points on top left are Newham and Brent, which both have equal sized white British and white other populations. The result of adding 'white Other' is to compress the graph substantially, as places like Bromley, Richmond, Kingston, don't have significant white other populations.

The next graph shows the same plot vs % Asian. This shows little correlation.

I dont understand that graph. What is it showing?

news.usc.edu/168987/antibody-testing-results-covid-19-infections-los-angeles-county

Investigating undetected c19 cases

God all we need is a Chinese vaccine! I assume it wouldn’t be given for free? Sounds a very handy way to make a huge amount of money. I really really hope we can find our own vaccine and support our own medical and manufacturing companies.

That London borough chart reads like a graph of more deprived to affluent.

If the Chinese vaccine is as good as their data and test swabs let’s develop our own please

whatsnext2 The first cases appeared in Italy on 31 January

Yes, it is possible that one or two people around that time went from China or Italy to the USA or elsewhere before they became ill

Someone could be infected but not pass it on if they had little contact with other people
Anyway, we've seen most people on ships didn't get infected when exposed together for weeks

You seem convinced that these infected people spread the infection so that it grew exponentially to give herd immunity to a significant % of the population of the UK, USA and other countries

.... all without being noticed in hospitals or in the death rates of these countries

That's the problem
There is no evidence of this exponential growth

If there were, we should see it as early epicentres of infection separate to the others
as well as in the death rates, if there were a sufficient number of cases to significantly increase herd immunity

However, in the USA, we can see that NYC was the early infection hotspot
We can see that London was the epicentre in the UK

It is possible the antibody studies may find far higher levels of immunity than predicted

but imo that is much more likely to be because COVID has a higher than estimated R0

• a recent CDC paper estimated it could be 5.7 -

together with a much higher level of infected people who never noticed any symptoms

So that the last 5-6 weeks has infected far more people than realised - from the known epicentres

Does the UK now have more deaths than Spain and Italy (around 22,000 and Italy around 25,000) or am I misunderstanding?

Here's a graph showing death rate vs. black population. There's not really much correlation.

It does seem that there is a reasonably straightforward correlation between having a high minority population and a higher risk of covid-19, but that this is not particularly associated with any particular minority.

However there are other factors - why are there relatively few deaths in Camden, and why does Barnet do so much worse than say Kingston?

Sorry I misread it. Not far off tho....

Aren’t Kingston and Camden particularly affluent? Newham is particularly deprived.

Deprived areas = worse health & higher population density ?

Pmk

Worse housing. Overcrowding .Possibly also much more likely to be going out to work. Having to shop regularly rather than big online orders.

More infection = higher death rate

So depressing. Elderly, vulnerable (health) and poverty all massive risk factors

@frompcat the Y-axis represents the number of deaths in the borough divided by the number of deaths across England & Wales by age averaged out across the age & population profile of the borough. In other words, if the figure is 100%, then there are the same number of deaths in that borough as you would expect across England & Wales, based on the age profile of that borough.

In fact, there are 2.7x the number of deaths across London (270%) as in England & Wales on average, adjusted for age profile, which isn't particularly surprising as you expect more deaths in a city.

That makes, the death figure for say, Kingston (113%/1.13x), incredibly low, as you have roughly the same death total as a random place in England & Wales.

(Sorry missed the graph of my last post)

@BigChocFrenzy there isn't really a correlation between population density, and death rate. Barnet has only 1/3 of the population density of Kensington & Chelsea, but a death rate which is 50% higher (300% vs 200% of E&W average). K&C has fourth-highest population density in London.

There are some geographical trends, e.g., going anti-lockwise around outer London: Hillingdon - 230%, Hounslow - 224% Richmond - 130%, Kingston - 113%, Sutton - 166%, Croydon - 273%, Bromley - 152%, Bexley - 160%, Havering - 197%, Redbridge - 243%, Waltham Forest - 283%, Enfield - 258%, Barnet - 291%, Harrow 334%

Ah, I see, Shoots I don't know the London boroughs

Population density does seem to play a big role in countries,
but not in cities, maybe because of diminishing returns:

everybody has so many contacts in their daily lives that being squeezed a bit further together doesn't increase the numbers significantly

@BigChocFrenzy "You seem convinced that these infected people spread the infection so that it grew exponentially to give herd immunity to a significant % of the population of the UK, USA and other countries"

I've never said that, in fact the opposite - adaptive immunity seems to be quite low with C19

Please don't put words in my mouth.

Then sorry, whatsnext I've obviously misunderstood you

I tthought you meant that these undetected early infections would mean we had a much higher level of immunity than estimates from Imperial and some public health authorities

So far we see levels of 3-4 % in a few places that have tested, with much higher levels in epicentres like NYC (20%), Gangelt (15%)

Anyway, the antibody studies should supply us with actual data instead of pointless speculation
The interim report in Germany will be sometime next month

I have no date for the UK report ?

@ShootsFruitAndLeaves. I know you have normalised for age but do you have a graph showing London boroughs with the oldest populations to eldest populations?

Kingston has a university, hence lots of students who are younger and don’t commute into central London daily. Lots of leafy suburbs appear to have better rates but also poor transport into central London so not everyone will be commuting into central London.

Although iirc PHE have told us people can catch COVID again, most opinion seems to be that this would only happen occasionally
but I have not read any estimates for how occasionally

Here's charts showing relative death rates for Wales, NW England and NE England

Lewis Goodall@lewisgoodall (BBC Newsnight)

I've been leaked #covid199^ death figures for Durham County Council for the 25 days to April 20th.

Nearly 50% of those deaths were in care homes.

72 hospital
67 care home
6 at home

If replicated across the country, it suggests our true death toll could be yet higher.

Confirms work @FT colleagues have been doing

And puts Britain pretty much bang into line with other European countries
where on average between 40%-50% of deaths have been in residential care.

Yes, the data from every other comparable country has around half the deaths occurring in care homes. Only in the UK to the government and scientific advisers keep up this pretence that 90% occur in hospitals