Daily numbers, graphs, analysis thread 5

[Barracker]

Welcome to thread 5 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
Google mobility stats

Thank you to all contributors for their factual, data driven, and civil discussions.

www.linkiesta.it/2020/04/coronavirus-morti-anziani-rsa/

Saw this on a site regarding nursing homes. I will translate.

Out of a total of 4629 nursing homes in Italy, of which 2166 were contacted by the Italian NHS, 577 answered the questions, 24% of the total facilities in the country.
According to the research, out of 44,457 residents (in 572 facilities [omissis]) up to the date of completion of the questionnaire on 23 March [omissis], there were 3,859 deaths, 8.4% (of care home residents).

So 8.4% of all nursing home residents seem to be dying in Italy looking at that data.

@Pluckedpencil

If those nursing homes are representative (and that’s a big if, as you might be more likely to report if you’ve experienced deaths), that gives 30-40,000 deaths! Of course, not all of these would be CV.

I agree with pp that blood donors would exclude people who have symptoms on the day,
but that would only be a % of the very small % who actually have COVID on that day

The Oxford results were far higher than the estimates of 3-6% from Imperial gave
and also much higher than studies and estimates in other European countries,
but the actual evidence for Oxford being so different was not convincing

I also disregard studies that have an obvious preference for a particular outcome or pet theory,
because even for very good scientists this can often unconsciously bias the study outcome

We should be aware - and hence be cautious - that the Oxford study was immediately pushed by certain political activists & business interests
who always opposed the lockdown and wish to end it immediately

I agree we need to end lockdown as soon as possible to rescue the economy - but not before then, not under pressure of very optimistic estimates

I would very much like references to any other peer-reviewed studies within the UK,
but until then, we can only go with what is available.

Anyway, within a few weeks, several countries will hopefully have a reliable antibody test and can the carry out a large test program of representative population samples

• but maybe not the UK ! -

to produce a reasonable first estimate of immunity rates and also how quickly they are increasing.

I think most people expected the Oxford study was well out. 3% is so low though. The one good thing is it does mean the Dutch mortality rate is only 0.65%.

As I outlined in recent posts, I think that understates the mortality rate I’m afraid. The 3-4 time lag between infection and death is a significant factor.

Agree re lag - I've essentially assumed that the same lag effects are present in all figures (e.g. Iceland's) which is of course a huge simplification but...effectively I've compared it to the rough way it's generally being calculated if that makes sense.

They announced that in the Republic of Ireland 55% of all deaths occurred in care homes
Yes, that would match the European figures showing that around half the coronavirus deaths occur in care homes. Meanwhile our British 'experts' and government continue to claim that it is only 10% here. Somehow we are different to every other country?

We should be aware - and hence be cautious - that the Oxford study was immediately pushed by certain political activists & business interests
who always opposed the lockdown and wish to end it immediately

The very sharp increase in deaths over past couple of weeks, in particular the extremely marked ONS spike, surely blows out of the water the Oxford study’s hypothesis that CV has been endemic here since January.

derby The RKI warned weeks ago that deaths would rise rapidly in Germany because there were so many cases that the track & trace teams couldn't cover all the contacts

So, far more infected cases were being missed and not isolated in time, or indeed at all if they had no symptoms.
Hence the spread increased for weeks, until finally lockdown worked and new cases fell again over the last week or so

The Imperial estimate (probably a bit low) of 0.75% infection rate would mean 600,000 cases instead of the 135,000 currently known

Track & trace is still vital for the UK to do as well though, because even when the system can't cover all cases,
it still reduces the gradient and the level of the later plateau

Sadly, I think we have only seen the beginning of deaths in nursing homes.

Whilst social distancing is lowering cases in the general population, nursing home cases can only accelerate in confined spaces.

@Derbygerbil If those nursing homes are representative (and that’s a big if, as you might be more likely to report if you’ve experienced deaths), that gives 30-40,000 deaths! Of course, not all of these would be CV.

According to the article, 37.4% (1143 out of 3859) of deaths displayed Covid-19 symptoms (testing positive (133) or displaying flu-like symptoms (1310)).

@BigChocFrenzy

Yes, I accept, that’s another important reason for the particular rise in Germany, alongside the impact of the infection-death lag.

Yes, if CV had been endemic for months, we should have seen a much earlier spike in deaths,
even if the cases themselves had been overlooked

There have been a few nasty bugs around this winter, but they were not COVID and not anywhere near the lethality of COVID

glass Yes, the outlook for nursing homes is bleak
and especially miserable if they must now also be denied visitors, for the sake of other residents and staff, if not for themselves

I agree re the carehome deaths, sadly I think there are many more to come. The population is enclosed and hugely vulnerable and the staff have limited poor quality PPE and in many cases inadequate training/experience to cope with this (not casting aspersions on the staff). Care agencies round here are crying out for staff too which will only make it worse.

We may well have the r0 well below zero in the general population but it's likely to be considerably higher in care homes. I'd also expect the level of infection amongst NHS to be hugely higher than any of the estimates.

Absolutely @BigChocFrenzy - and the lack of PPE/testing.

There are also lots of vulnerable elderly in the community, who have a social care package of up to 4x carers daily, do their risk is huge.

So sad.

I'm an NHS worker (thankfully working remotely) & the lack of staff testing is shocking too.

I cannot see there being 100K rests daily by the end of the month; in my workplace you need to be a critical worker (ie your team cannot function without you) PLUS have authorisation from a very senior manager PLUS be in the 1st 72hrs of symptoms. So very few staff are tested.

I have friends/family working as paramedics & in ICU who have not been tested. It's outrageous.

We may well have the r0 well below zero in the general population but it's likely to be considerably higher in care homes. I'd also expect the level of infection amongst NHS to be hugely higher than any of the estimates.

It would be really interesting and useful to know where new infections are coming from in Italy. Are these mainly in their health service and care homes? That would indicate where weaknesses were in our CV defence that we could address.

@GlassOfProsecco

I agree... completely outrageous.

I believe the 72 hour limit is due to the efficacy of the test.... Any longer and it is no longer reliable, which is all the more reason for NHS and care staff to be tested swiftly on demand.

Sorry below 1, r0 below zero would be miraculous!

Even people going to hospital aren't always being tested here - friend (in her 60s) has been laid up with covid symptoms for 2 weeks, really really unwell. Eventually told to go to hospital to get checked. Told she almost certainly had it and that she also had a chest infection, and sent home with antibiotics. No test done.

• • • DAILY UPDATE * * * Thursday APRIL 16th

Total UK cases: 103,093
New UK cases: 4,618
Total UK Deaths: 13,729
New UK Deaths: 861

Today's volcano:

Thank you Barracker!

Applause for the volcano - always the high spot of my evening
< I must be a sad geek ! >

Thanks, Barracker
It will be interesting to see how the Easter holiday period fills out with more data

John Burn-Murdoch@jburnmurdoch (FT stats geek)

Colour = phase of outbreak
• Red: more new hospitalisations every day than day before
• Orange: total hospitalisations rising, but rate of increase slowing
• Blue: fewer people in hospital than before
.....

UK:

• London’s hospitals now have fewer covid patients with each day
. Rate of increase slowing elsewhere across UK

• Suggests UK may be at or near peak for new infections
< but not necessarily past peak deaths >

Italy:

Almost all Italian regions now in "reduction" phase :-)

Spain:

• Madrid now firmly in the "reduction" phase

• But in most other regions, although the rate of increase has slowed, each day still brings more new hospitalisations than discharges

France:

• Rate of increase in hospitalisations dropping across the board, but most regions yet to see total occupancy fall