Daily numbers, graphs, analysis thread 10

[BigChocFrenzy]

Welcome to thread 10 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

We welcome factual, data driven, and civil discussions from all contributors 💐

I think whether it will hit 500k in uk will depend on the t-cell and antibodies and how long immunity lasts etc. We've got maybe 5.5% immunity for 40k dead.

Unless there’s stuff I’ve missed, t-cell immunity is the great unknown that could make a significant impact on how manage Covid, and how deadly it is.

The question I have does the effectiveness of t-cells depend on viral load, in a way that antibodies wouldn’t be. In other words, if viral load is low could a healthy person with effective t-cells ward off Covid before an infection takes hold, remaining asymptomatic.... but become ill if viral load were high, and require antibody production. If so, T cells provide some protection, rather than make you immune, from Covid.

German virologists say most measures can be relaxed while probably avoiding any 2nd wave,

but the key is keeping the ban on superspreader events and reducing the number of people a superspreader can infect
... based on analyses like this:

https://www.sciencemag.org/news/2020/05/why-do-some-covid-19-patients-infect-many-others-whereas-most-don-t-spread-virus-alll^

SARS-CoV-2 .....Without social distancing, this reproduction number (R) is about three.
But in real life, some people infect many others and others don’t spread the disease at all.^
.....
“Probably about 10% of cases lead to 80% of the spread,”

T-cells may be a major reason for the great difference in severity of this disease between the very elderly, who mostly don't seem able to use this mechanism
and young adults who often (?) may use T cells, rather than antibodies

When we get national serology studies of the UK and a few other countries,
I hope there will be full breakdowns of age especially, also sex, race for both antibodies and T cells

Peacemaking!

T cells seem to be cropping up as the explanation as to why some people have very mild or no symptoms when infected. But if this was the case, you would expect virtually no asymptomatic infection in the elderly... wouldn't you?

Whereas, in fact, lots of asymptomatic cases are popping up when blanket screens of nursing homes are done - frail 90 year olds who are just fine, but whose tests are coming back positive. How are these people fighting the infection? I'm absolutely not an immunologist, but everything I'm reading seems to suggest that T cells have left the building by then. Such a strange illness.

New fred, new graphs

Here's the weekly deaths in homes/hospitals/care homes to week 22 (death occurred in week to 29 May, registered to 6th June)

I've included the week 19 graph to show the extra deaths that will be registered within the next few days and weeks. That's +11% for home deaths, +5% each for hospital and care home deaths.

There was a massive fall in care home deaths this week to 2406 from 3134. This means deaths are now well below normal January levels, even after adding 5%.

Deaths at home remain relatively high, at normal post-Christmas levels. It's not clear however how much of this is due to hospital being almost totally empty, with further falls in the number of people dying in hospital to far below any previous week this year (no word on previous years).

The headline 'registrations' count is just 9824, however this was a bank holiday week and relatively unworthy of comment or discussion.

The total number of deaths is likely to end up in the 10,500 to 11,000 range for the week ending 29 May 2020 by date of death (noting that we are so far missing all the car accidents and suicides). This is more than usual deaths for May but very far from pandemic levels

I.e. I believe the ONS erred in their data categories.

They provide these venues of death:

• home [defined as the usual residence of the deceased]
• hospital
• care home
• hospice
• other communal establishment [hospitals, care homes and hospices are also communal establishments, but so too are children's homes and prisons]
• elsewhere [this would be someone being squashed by a truck and dead on the road for example, or being killed in a hotel by their boyfriend]

It would make more sense I think to separate the 'hospital' stats by normal place of residence. So 'hospital - care home resident' and 'hospital - non care home resident'.

this would be someone being squashed by a truck and dead on the road for example, or being killed in a hotel by their boyfriend

you feeling ok shoots ? Been watching a lot of Scandi Noir ?

we need you to remain ok as so many of us are relying on your intepretations and signposting .

"Deaths at home remain relatively high"

UK coronavirus victims have lain undetected at home for two weeks

A natural consequence during lockdown when some elderly live alone, have no close family and take advice to stay home as much as possible

(tbh, one minor reason I - as an Aspie hermit - moved to a serviced flat, is so that my eventual corpse won't lie undiscovered until neighbours complain about the smell)

https://www.theguardian.com/world/2020/jun/07/uk-coronavirus-victims-have-lain-undetected-at-home-for-two-weeks

Thanks Shoots.

They updated their death by setting graph too. Deaths down in all settings, but care homes making a larger proportion with the decreased overall count.

The pure care home data are here

www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/datasets/numberofdeathsincarehomesnotifiedtothecarequalitycommissionengland

.....

but they don't match up to the death registrations.

It can be seen from Table 4, that the number of non-hospital covid deaths by care home residents is relatively constant at around 200-250 each week

This is less than 10% of normal weekly care home deaths, which seems far too low, and suggests NHS hospital treatment is being denied to care home residents.

Actually looking at the last week, care homes and others are a smaller proportion.

"It would make more sense I think to separate the 'hospital' stats by normal place of residence. So 'hospital - care home resident' and 'hospital - non care home resident'."

Yes, so we can get a complete picture of the % dead who are / were care home residents

Currently, many people are still saying the dead would have died within a few months anyway / should have died last winter but it was mild

whereas other estimates are that the average lost life expectancy is 10 years

Obviously care home residents have a much lower life expectancy than others of the same age living independently, so we need to know the proportions

At age 80, the average woman has a life expectancy of 10 years, a man 9 years

www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/healthandlifeexpectancies/articles/lifeexpectancycalculator/2019-06-07

However, this ONS calculator presumably takes a population average, but we know that care home residents have a higher risk of death,
so this average probably is not suitable for COVID calc

but we must also consider the lost years for those who die under say age 75
and especially the small minority of working age, < 65

"care homes making a larger proportion with the decreased overall count"

showing that infection within the community is at a much, much lower level than before

Thanks @peridito. Not sure what Scandi noir is but it sounds interesting. I just watched Gone Girl which I thought I'd already watched but it turned out I hadn't. It has sometimes occurred to me while watching TV/movies that 'that wouldn't work during a pandemic' about some everyday seen of normal behaviour. I don't know if covid-19 will be written into popular entertainment.

I did watch 'Contagion' recently and it was really meh because it was like this sort of ebola instant death thing when actually we've found that covid-19 is rather more mundane than that.

I had a brief look for SARS (2003) deaths by age but didn't find it. I know the CFR there is 10%+ but wonder if the IFR actually much less terrifying, or it really was that lethal for infected people. (depends how many were tested.)

At peak, care homes were about 30% of deaths, but the absolute numbers are much lower, even though they have not dropped proportionately as much as in the community.

Care homes also reportedly have fewer residents now, which would help with infection control

• not just deaths, but people avoiding care homes when possible

SARS indeed had 11% IFR, but only about 8,500 cases worldwide

R0 was 2-4, but it was relatively easy to reduce to a v low R0 once preventative measures were established

iirc, it mostly arose in hospitals among sick people and medical staff,
so effective infection control / barrier nursing was mostly what zapped it.

It was much easier to tackle than COVID - partly because the IFR being 10 x higher meant it was recognised as an emergency v early on
and that is when measures against an epidemic are most effective.

CFR, not IFR

"worldwide" - apart from an outbreak in Canada, it was mostly confined to a few countries near China

Placemarking to catch up. Hope Barracker is ok

Thanks BigChoc for the new thread.

Yes, Case fatality Rate 11%
iirc, Mild / asymptomatic cases reportedly can give some immunity to COVID via T cells ?

Do asymptomatic people develop antibodies?

It depends what you call asymptomatic. Showing enough symptoms to justify testing or feeling a bit off? I've said this before; none of dd's nursing friends were ever diagnosed or even tested for Covid, all but one have tested positive for antibodies.

Having said that, in order to be considered to have symptoms back in March they had to have a high temperature for 48 hours and a persistent cough. Many of them had loss of taste and various other minor symptoms.

We should surely expect lower than normal deaths in care homes for the rest of the year shouldn't we? Working on the premise that many of the most vulnerable may have died.We won't probably know until this time next year whether the "most would have died soon anyway" narrative is true.

Yes, we should expect lower deaths just because there are 10%+ fewer residents than normal due to both excess deaths and deferred replacement due to lockdown. And probably lower deaths for quite some time in that all care home residents are vulnerable so they are all at risk of death from covid-19 so we will have killed people who would have otherwise died in 18 months time or whatever.

It is a given that they would have died relatively soon in that life expectancy for care home residents on admission is about 2.3 years, but January tends to kill a lot .

I imagine the average new patient is unlikely to die imminently, so the cohort of new patients will nearly all survive winter 2020-2021. Perhaps a death spike in 2022-2023?

I had a look at some of the SARS reports. It does seem that testing was a bit shit (non-existent during much of the pandemic) and the definition was having clinical symptoms until they had a test. So probably IFR was much lower than CFR, but I can't find any reasonable estimate on what it was.