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.

@oralengineer

I agree that some of those least at risk are most risk averse. Lots of people are really bad at assessing risk and responding to that assessment in a reasonable manner.

However, I don’t think it’s as straightforward as that in many cases. In the same way that the risk is far, far less than 1% for many people, it is also greater than 1% for others.

Put simplistically, and with numbers below being illustrative only, a 1% death rate is equivalent to 90% being very low risk, say 0.01%, and 10% being higher risk, say 10%.

Much of the 90% will have close family or friends who are in the 10%, and perhaps a quarter will live with them.... so even if they’re not particularly at risk, if they catch it, they risk passing it on if they don’t refrain from social interactions. And unless you’re a sociopath, you’re anxious for your actions not to be responsible for Covid to play Russian roulette (which a 10% is roughly akin to) with your vulnerable close family and friends...

Derby
COVID has been running rampant and unchecked in this country for several months
lockdown has reduced transmission
but random testing shows less than 10% have been exposed
so 90% have not
and unless they have caught it in the last few days or are ill
do not have it
so letting lots of people who do not have a disease meet up
will not magic it up

north Devon has had no deaths and sod all cases (all in nursing homes)
how exactly would it magic up in the schools if they opened?

ditto 90% of the country

re the SAGE concern about R0 if schools reopen:

We had the same concern in Germany and the national rate was indeed calculated to be briefly above 1.0
However, the new cases were isolated outbreaks in institutions or places of work, which could be fairly easily locked down locally and R0 has since fallen well below 1

What helped ease worries before school return:

• Several weeks of discussion between schools and local authorities, with detailled plans signed off
• Both kids & staff can wear masks if they wish
• Plenty of space in classrooms because of half-classes
• Plenty of sinks & loos - but afaik that was always the case
• Almost all schools in Germany are normally only 8am - 1 or 2 pm, i.e. no lunch

Schools break up anyway at end June, but these weeks are a useful transition back into education

I expect schools here will be ft after the vacation

But hadn’t appreciated they didn’t include Carr homes. However, given that only 4% of over 65s live in care homes, their omission may not distort the figures that much...... (though appreciate that % will rise in over 80-90

A massive distortion.

Around half of all excess deaths are in care homes. This is 22,500 people. So 22,500 out of 400k. At 400,000 residents, 10% infection fatality rate, infection rate would be more than 50% of all residents

If you have one population of elderly with 5% infection, and another with 50%, then you aren't going to get meaningful numbers.

Half of dead in care homes. So if we consider perhaps only 20,000 deaths acquired in the community, then the IFR looks much lower....

north Devon has had no deaths and sod all cases (all in nursing homes)

Don't post things that are not true. There are deaths. Check the official stats before posting.

Shoots
My error. I looked at the ward by ward data.
10 had not had a death, one had.

In my area I can identify every death from my facebook feed

@ShootsFruitandLeaves

I accept that the infection rate in care homes has been disproportionately high, and that this would need to be normalised for a population-wise IFR to be calculated. However, a population scale IFR is of limited use expect to benchmark against other diseases and give an indicator or relative severity. You can get wildly different IFRs if you choose to exclude certain groups, but on the basis that care home residents are as much human as the rest of us, it feels wrong to exclude them if trying to determine a generalised IFR.

Also, my understanding is that hospital deaths to today were 30,376, though I accept some of these would have been admissions from care homes.

How many pupils and teachers are shielded or living with a shielded person ?

It should be possible to technically support shielded teachers to support shielded children / families
while letting the great majority of pupils return

The majority of conditions classed as merely "vulnerable", e.g. T1 or BMI 40+ seem only to bring high risk for those above normal working age

With about 15 million vulnerable, millions will inevitably have to work outside the home, if they are not alteady doing so.

Possibly 100,000s of shielded / with shielded family members will have to as well,
to keep their jobs and pay the bills

• this group is concerning, as we don't know enough about risks for the various comorbidities & combinations at different ages up to 65.

Presumably most of this group are shielding atm, so statistics to date may underestimate their risk.

but on the basis that care home residents are as much human as the rest of us, it feels wrong to exclude them if trying to determine a generalised IFR

Indeed

That's why your quoting 5% infection rates is irrelevant when half of total deaths are in care homes

@ListeningQuietly

I’m not sure you’ve understood what I was trying to say... I recognise that only a minority are believed to have been infected in the UK (7% according to recent ONS estimate).

My use of 10% low key risk / 90% high risk was merely an illustration to attempt to show that it was rational to be fearful about the personal impact of Covid-19 even if you were individually in a very low risk group. As I tried to make clear, the 10% and 90% were only chosen as figures to aid the illustration - they have actual basis apart from being very rough approximations for those in high and low risk groups.

That's why your quoting 5% infection rates is irrelevant when half of total deaths are in care homes

I‘m not sure where the 5% comes from - it’s not a figure I’ve used. Also, I accept care home deaths are high, but are they 50%? The ONS has 10,636 in care homes compared to 30,000+ in hospitals.

www.ons.gov.uk/file?uri=%2fpeoplepopulationandcommunity%2fbirthsdeathsandmarriages%2fdeaths%2fdatasets%2fnumberofdeathsincarehomesnotifiedtothecarequalitycommissionengland%2f2020/20200524officialsensitivecoviddeathnotificationsdata20200522.xlsx

listening NORTHERN DEVON HEALTHCARE NHS TRUST lists only 21 deaths from COVID so far

https://www.england.nhs.uk/statistics/wp-content/uploads/sites/2/2020/05/COVID-19-total-announced-deaths-28-May-2020.xlsx

That spreadsheet starts in mid-April

Check excess deaths for care homes

teachers are mainly concerned for their own welfare

How very dare they.

Excess deaths in care homes in April account for at least 18,000 of the 60,000 extra deaths. As far as I can see there is no May care home data published yet.

www.bbc.co.uk/news/health-52284281

I think it is fair to remove these numbers when discussing the risk to school children and their families, but not when looking at the IFR for the country as a whole, this is because the risk of dying of covid is much, much, much greater if you are in a care home rather than in the general population.

So, 18,000/400,000 x 100% = 4.5% chance of dying if you live in a care home of COVID or as a result of COVID before May 1. (This number will increase when May numbers known).

42,000/59,000,000 x100% = 0.07% of dying of covid outside of a care home (this number will decrease once May care home deaths are included.

The difference would be even greater if we subtracted May care home excess deaths from the figures.

For individual risk of dying, the overall rate is not helpful, e.g. an 8 yr-old vs 80-yr-old

Instead, look at:

• your age - absolutely dominant
• your sex - much higher risk at the same age for males
• any significant health conditions

Imo d&c are at very high risk - once they start ketting people stay overnight.
Wsm already getting more cases. And because very little immunity it will spread quickly locally in the supermarket and shops

@Keepdistance there is little immunity everywhere? Slightly if her in London, but still very low.

Excess deaths in care homes in April account for at least 18,000 of the 60,000 extra deaths. As far as I can see there is no May care home data published yet.

I posted them earlier in the thread.

Don't use the BBC as a source. They can't count. Check the ONS.

www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/causesofdeath/datasets/deathregistrationsandoccurrencesbylocalauthorityandhealthboard

I’m still exercised as to what constitutes immunity. We know that the virus is easily transmitted, but not how infective it is. It seems to me that in care homes full of people vulnerable due to age, with presumably numerous comorbidities, and expected to die sooner than the general population, and being cared for by individuals lacking in proper PPE and possibly training and experience, one would have expected more infections and deaths than have actually been seen. I think there has been some surprise as to the low levels of infection in a number of populations. Maybe the vast majority of people simply don’t get it, even if exposed?

without antibody tests we dont know what level of immunity exists in children. In schools that had half term skiing trips to Italy I suspect there will have been high levels of asymptomatic infections.

Did your child have Covid toes? If mine was school age I'd ask the school to check with parents if any did.

A delayed apology @whatsnext . I've seen that I missed your comment about the elderly in ICU and repeated the point!

We went to Italy feb half term sking. Dd was very ill two weeks later but tested negative.

However she had
Fever
Muscle pain
Headache
Sore throat
She appeared to pick up then 3 days later got
Fever again
Cough
Hoarse voice for a week

I and DH got coughs tiredness and upper back pain. I had a 3 day headache.

My son had a migraine and got covid toes. His mouth was full of ulcers. And then I had transient hives.

I'm desperate for a antibody test.

Certainly without antibody tests we don’t know what level of acquired immunity exists in children. We also don’t know how long that immunity lasts for, at present.
Also, given the hygiene habits of children and their propensity to touch and lick things, I suppose it’s possible for a child to be a vector of disease without actually contracting it. I think it was on this thread somebody mentioned one country where the main plank of lockdown-exit precautions is hand washing.

@ShootsFruitAndLeaves

I was also looking at this ONS page for deaths of carehome residents as it counts both deaths in care homes itself (9,039) and in hospital (3,444) of care home residents with COVID on death registration.
www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/articles/deathsinvolvingcovid19inthecaresectorenglandandwales/deathsoccurringupto1may2020andregisteredupto9may2020provisional

It doesn’t have compare excess deaths against the 5-year average, however, only Against 2019. It suggests this is up 23,136, year on year. (Lots of Alzheimer’s and dementia).

I think this report needs to be read alongside the FT excess death count to see how significant the care home deaths have been as a proportion of the 60,000 deaths comparing different countries that makes headlines.