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.

Ok thanks shoots that helps clarify my thinking and my assumptions. I also vaguely remember the ons saying they were going to look into the non assigned covid deaths at some point, so that should help.

On the difference between England and other U.K. countries on death rates my memory is that NI effectively had earlier school and business closures because of the St Patrick’s day holiday on 17 March.
The catholic bishops who control quite a lot of schools in NI advised their schools to take discretionary holidays and close from March 13. They never reopened.
A lot of businesses closed on Monday 16 March because Tuesday 17 March was a holiday. People avoiding going back to work and wfh after that where possible.

Thanks innu for the reply.
For wales maybe they did not have a policy of emptying hospitals in the same way ? Just surmising - no idea ,will do some digging.or different policy of testing - though as @ cathy says testing not great in wales. Thanks for the help everyone.

Discusses regional variations and excess deaths:
www.medrxiv.org/content/10.1101/2020.05.26.20113357v1

Use of steroids decreased mortality very significantly:

www.medrxiv.org/content/10.1101/2020.05.22.20110544v1

Thanks @ what, will have a read.

@Humphriescushion
Initial hotspot in Wales was SE, centred on Royal Gwent Hospital, but it was about a week or two behind the London peak. Highest death rate overall though has been the Cwm Taf Health board ( includes Rhondda and other valleys) where socio-economic factors have likely affected outcomes. There is still a bit of a hotspot along the A55 corridor in North Wales but in rural West and Mid Wales, cases/deaths have been very low.
I commute from West Wales to Cardiff to work and the entire 3-week period in March before lockdown, I was usually the only person in the train carriage.
So a combo of behavioral change ahead of lockdown and formal lockdown happening at a more optimum time following initial seeding probably helped Wales' numbers.
HTH

Ah thanks @ blody that makes sense - and i think i have read somewhere that late lockdown has lead to the high deaths in the uk and if wales was behind that makes sense. Any connection to market hall cinema in a certain place?

That steroid study is interesting but patients were not randomised to treatment condition - I've just had a quick read and I couldn't really tell how they were allocated, so the results need to be interpreted very carefully - I doubt it's the miracle treatment that it might seem at first glance.

North Wales - Betsi Cadwaladr hospital area - hadn't reached peak and was still increasing last time I looked. Rural areas had few cases but the deprived communities along the coast had more.

Not much PT in lots of areas of Wales. The hotspots being in the south so connections to airport and Bristol/london etc and the Concert.
Also like Devon an initital rise in numbers meaning people more careful later on.
Personally think tower blocks more likely to be an issue and there are fewer of those in Bristol/wales/devon etc. But many in nyc etc.

Encouraging French research about antibody formation in mild cases. 97% mounted a response after 4 weeks. Here's a Guardian link- off to hunt down the original paper...

www.theguardian.com/world/2020/may/27/french-team-finds-mild-coronavirus-infection-does-lead-to-antibodies

Here it is: www.medrxiv.org/content/10.1101/2020.05.19.20101832v2

Apologies if this has been discussed before, but has any consideration been given to the comparatively much worse number of deaths in England than in other countries of the UK? I must say being in England I’m not aware of the exact lockdown rules in other countries, but I wasn’t aware that within the UK they had varied so much to cause the change?

The data I’ve seen is at the bottom of this website - www.travellingtabby.com/scotland-coronavirus-tracker/ - and shows england at 596 deaths per million compared to Scotland’s 424, Wales’ 412 and NI’s 274.

I also noticed that in Scotland there have been 23 deaths in those under 45 and no children have lost their lives. I believe this is comparatively lower than the UK as a whole. I’m not someone who subscribes to the idea that lives of young people are in any way of more value than the elderly (indeed I’m old enough to be considered collateral damage by many!), or that fewer deaths of young people at the expensive of the elderly is in any way a sign of ‘success’. However, it does suggest perhaps lower rates of transmission outside of health and care settings, or perhaps better outcomes for younger people admitted to hospital.

As I said, this may have been discussed and I could be stating the obvious! But it struck me that all this talk of the UK’s outcomes don’t necessarily reflect the full picture and thought there may be some way of an explanation?

England has significantly higher population density than Wales, Scotland and Northern Ireland.

Almost one-third of the population lives in the southeast and 9,000,000 in London. The large urban centres ( London, West Midlands, Manchester, Leeds) saw the worst of the outbreak, particularly in the early days.

Is population density not a bit of a red herring though? In Scotland around three quarters of the population live in the central belt, not quite the same as London I appreciate but equally it isn’t universally sparsely populated (just as not all of England is densely populated). I appreciate it is a consideration, but I am sceptical as to how far it explains things.

But in general the more densely populated areas have fared worse. London is also a major international hub and I think packed public transport can facilitate spread.

I wonder also whether the lockdown was effectively 'earlier' in the other parts of the UK in terms of number of infections / point in the curve - and thus demonstrate how much better England's outcomes might have been with a lockdown earlier in the curve?

Yes- but now there are rising cases in the NE, North Wales and other areas that were behind the curve. hopefully they won't be anywhere as near as bad as the London/West Mids peaks though.

FatRascalsAndJam I don't believe it is a red herring. I think if you put density together with BAME population, you potentially could have the reason. Scotland has a very low BAME population, googling away it looks like 5% v. 20%. Add to that the higher BAME death rate, which is being analysed, plus the population density and you may have your answer.

Age distribution is roughly the same, and care home deaths proportionately seem to be higher in Scotland, so that's broadly similar.

Great thread! I've been lurking for a while...

I just wanted to say that I don't think we can assume that most of the excess community deaths are from COVID-19.

I'm a GP working in an area with high levels of deprivation, and many of my patients are too scared to attend hospital under ANY circumstances at present. We're having more deaths than usual for the time of year, but only about 2/3rds were COVID-19 symptomatic, and these include the nursing home patients who had EHCPs agreed not for admission.

There are, sadly, many people sitting at home with their heart attacks and strokes, not seeking help, and only calling 999 when they realise they're dying. This is gradually improving, but many of the excess deaths will be non-COVID.

I would consider those deaths to be indirect Covid deaths. They happened as a result of the lockdown.

FatRascalsAndJam I don't believe it is a red herring. I think if you put density together with BAME population, you potentially could have the reason. Scotland has a very low BAME population, googling away it looks like 5% v. 20%. Add to that the higher BAME death rate, which is being analysed, plus the population density and you may have your answer.

1000 times NO.

BAME people are NOT dying in significant numbers.

84% of covid-19 deaths in England & Wales as of 10 April were white. This compared to 86% for the underlying population.

At this time covid-19 had hit London, which is majority BAME, VERY hard, and not so much the rest of the country.

In addition, many thousands of almost 100% white people were NOT being counted in care homes at that itme.

On the cut-off for the ONS survey, hospital deaths were the largest group. However, past that date, hospital deaths have fallen below normal (non-covid) date, while care home deaths have grown to make up a larger proportion.

There have been a further 15,000++ care home deaths since the ONS survey. These will have been 96% white. The number of hospital deaths since that date is around 5,000.

The typical covid-19 death is in fact a woman in her 90s. No, not a man. Women are much more likely to have survived to that age, and because AGE (not ethnicity) is the overwhelming predictor of covid-19 death (despite what the media would claim), this is the largest group of deaths, despite men having greater risk.

There is an exponential relationship (a straight line between the logarithm of age and death chance) between age and death risk. There are almost no BAME people who are 70+, relative to many millions of whites.

Don't forget that the white population of the UK has not grown since 1939, while the rest of the world has added billions of people. These people are too young to die in large proportion. It's literally impossible.

Report from the NYT saying that the first confirmed coronavirus infections in Italy and the USA didn't trigger the epidemics. They've looked at viral genomes, which suggest the wider spread began weeks later.

The new study suggests, the coronavirus arrived more than once without starting runaway outbreaks. In these cases, there was little or no transmission, and the virus simply died out.

www.nytimes.com/2020/05/27/health/coronavirus-spread-united-states.html?smtyp=cur&smid=tw-nythealth

"1000 times NO.
BAME people are NOT dying in significant numbers.

Exactly
For deaths, AGE is absolutely dominant and only a v small % of BAME people are very elderly

COVID is overwhelmingly killing v elderly white people

Young and middle-aged Black people and certain Asian groups may be at higher risk than whites of the same age - and we have to consider occupation risk distorting that too -
but very few of any colour are dying under age 45
(and almost no children)

London does have a much higher % infected obviously, because it's a world city with a huge number of international connections and an 8 million pop

However, because of its high % BAME - and hence younger - population,
London probably has a lower death rate of those infected than other areas of England