Daily numbers, graphs, analysis thread 11

[BigChocFrenzy]

Welcome to thread 11 of the daily updates

Resource links:

Slides & data UK govt pressers
NHS England stats including breakdown by Hospital Trust
ONS UK statistics for CV related deaths outside hospitals, released weekly each Tuesday
Financial Times Daily updates and graphs
HSJ Coronavirus updates
Worldometer UK page
Covidly.com to filter graphs using selected data filters ONS statistics for CV related deaths outside hospitals, released weekly each Tuesday
Plot COVID Graphs Our World in Data

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

covid.joinzoe.com/data

It's interesting because the percentage number of positive antibody tests across the uk is estimated at 5-7% but the drop from last week is 35 %.

Obviously I'm not understanding some key factor here!!

The purple animation seems to end in late May too

I don't find their data very informative tbh. It was nice watching the animation coming down a couple of months ago

They book you a test very quickly and easily though.

They're giving app users access to info they're sending the government, which is more informative.

told not to show anyone cases are rising by the govt?!

Hmmm.

The Zoe app appears to be saying -

• the number of people per day with positive tests plus those with positive antibody tests (call this X) was 4,900, then 3,612
• the number of people per day with positive tests (call this Y) is 2,341

X and Y are not comparable because of the change in data.

They say Y has stopped declining BUT are not telling us what Y was historically, so we have to take their word for that (and how fast it was falling, etc). So whilst it's good that the overall number of new infections seems to be a low actual number, they are raising caution that it's not declining (but providing no evidence for that).

This strikes me (and I fully admit I am the suspicious type) that errors have been made and, perhaps, they weren't fully aware of the inclusions of antibody tests (which would never have made any sense being included in numbers of "people acquiring the infection per day" basis)...

Regarding lung function. I did a course on COPD some years back and was struck by the fact that it is often diagnosed late as lung function can be as low as 50% of expected before patients feel symptoms and seek help.

If I've remembered that correctly it suggests that there could be a lot of people out there with undiagnosed lung problems (particularly current and ex-smokers) but, also suggests that problems after Covid may not be picked up unless severe. It could also be a reason why some people seek help so late when symptomatic

PatriciaHolm - yes, I think you are right. I initially thought that Zoe used the results of antibody tests to backtest and revise their estimates but it looks like the explanation might be simpler - there were errors made in their original calculations.

Taking London as an example - their last week estimate was 54-125 new cases per million of people per day. They have mentioned London as a potential area to watch. This week number has fallen down to 4-38. It is a very significant adjustment but it makes sense if Zoe indeed counted some positive antibody tests (which would be higher for London) in their daily estimates.

Tim Spector perhaps needs to listen to Radio 5's More or Less, they've been all over the testing figures and their presentation by the government!

Sorry, Prof Tim Spector is the scientist leading the Zoe app

According to Zoe there are currently 132 000 people with covid in the UK. Assuming 66 million population, that's 1 in 500.
How does that tie in with the briefing statements of 1 in 1700 people infected?

Zoe isnt a representative sample. I discussed this in the last thread. Too many white middle class people,.not enough older or younger

Shoots,

Zoe know their demographic and can adjust accordingly.

They are respected by a number of senior decision makers.

"sometimes the testing turns up things you would never have known about (and might never have harmed you) if you hadn't tested."

I know you just want to "get back to normal", but wilful denial of problems is not solving the problems

Scans of organs in some survivors has revealed damage that will be longterm
Some COVID survivors will have longterm effects, that may even be permanent
Denial won't change that

We need to know how many people this involves and make sure the resources are made available to treat them

• and that benefits are set up for some who will be unable to work in the interim.

That is completely independent of restarting schools and the economy,
so I don't see the point of denying things that have already happened

I couldn’t register my antibody test on Zoe, it only says ‘tests’ and didn’t discern between. Antibody or antigen. So my negative antibody result is counted as a negative antigen?

Tests suggest 5-7% on average across the uk have antibodies
Reports are that the number of antibodies drops sharply within months

==> Is this more than usual after a virus, or will there still be sufficient immunity for 1-2 years ?

An unknown number will have T Cells

==> Anyone have any estimations of what % ?
Seems unlikely in the elderly, but may be quite common among say the under 30s

Evidence for other Coronaviruses suggests that T cells can be protective for many years
==> will those pople still be able to get infected and pass on the infection ?

@nonamehere

According to Zoe there are currently 132 000 people with covid in the UK. Assuming 66 million population, that's 1 in 500. How does that tie in with the briefing statements of 1 in 1700 people infected?

They appear to have adjusted their daily infection rate because of the antibody issue, but not their estimate of overall infection. Probably because a) they can't, because they are not sure about the data and b)if would look very odd and ruin the timeline.

I am sceptical.

@Wunderweb

I couldn’t register my antibody test on Zoe, it only says ‘tests’ and didn’t discern between. Antibody or antigen. So my negative antibody result is counted as a negative antigen?

I think that's probably why they can't adjust any historical numbers. They haven't been asking the right questions.

I'm astonished if Zoe doesn't distinguish between types of tests
Combining that data without this info would mess up both

@BigChocFrenzy

"sometimes the testing turns up things you would never have known about (and might never have harmed you) if you hadn't tested."

I know you just want to "get back to normal", but wilful denial of problems is not solving the problems

Scans of organs in some survivors has revealed damage that will be longterm
Some COVID survivors will have longterm effects, that may even be permanent
Denial won't change that

We need to know how many people this involves and make sure the resources are made available to treat them

• and that benefits are set up for some who will be unable to work in the interim.

I'm not denying anything.

I'm pointing out that testing can reveal things that you didn't know about and that might never have harmed you.

It's a well known issue with cancer testing, particularly breast cancer - are women being treated for tumours that were discovered through testing that would never have harmed them when the treatment harms them?

Just because someone who has tested positive for asymptomatic Covid has lung damage, doesn't mean the virus that caused no symptoms caused the damage. That's hardly controversial.

So how do they isolate the cases where the damage was caused by the virus that didn't cause any other symptoms and the cases where the damage was caused by something else?

My opposition to lockdown as a public health measure doesn't render me unable to accept facts.

Zoe base some of their predictions on symptoms I believe? They now also include a range of other questions each week such as how often you've been out etc.

Facts are always interpreted as such via the eye of their beholder, on all sides of any debate.

Also, there's political decisions which may be different to medical decisions and also social decisions.

We can all interpret those differently depending on our own personal and wider social contexts.

Abattoirs:

Also highlights the risk of air con in other work places with a large amount of aerosol droplets
btw, I found that my gym has not yet switched on its new air con, despite >30C - the owner was already concerned about the risk

https://www.theguardian.com/world/2020/jun/25/abattoir-air-cooling-systems-could-pose-covid-19-risks-expert-warns

Air cooling systems used at abattoirs could be an overlooked risk factor accounting for Covid-19 outbreaks,
according to scientists who have studied conditions at a meat-processing plant at the heart of a cluster of infections in Germany.
....
At a press conference, Exner said the air filtration system in the slaughter area had contributed to the spread of aerosol droplets laden with the virus,
describing it as a “newly recognised risk factor”.

The area of the plant where animals are slaughtered, gutted and cut to pieces is being kept at a cool 6-10C degrees.
To do so, however, the cooling system merely circulated the same unfiltered air, thus keeping aerosols in motion...
A filter fitted to the cooling system was not able to keep out the virus