Daily numbers, graphs, analysis thread 13

[PatriciaHolm]

Taking the liberty of starting a new thread as we've just bust the old one, with much thanks to @BigChocfrenzy and I will copy her header..

Welcome to thread 13 of the daily updates

Resource links:

Slides & data UK govt pressers
UK dashboard sub-national data, local authorities
Beta Uk dashboard deaths, cases, hospitals, tests, partially sub-national
UK stats updated daily by PHE & DHSC
ONS UK statistics for CV related deaths, released weekly each Tuesday
PHE surveillance report infections & deaths released every Thursday with sep. infographic
NHS England stats including breakdown by Hospital Trust
FT Daily updates
HSJ Healthcare updates
Worldometer UK page
Plot FT graphs compare countries deaths, cases / million pop. / log / linear
Covidly.com filter graphs compare countries
Plot COVID Graphs Our World in Data

We welcome factual, data driven, and civil discussions from all contributors 📈📶👍

@LarkDescending yes that was an issue originally, but they amended for that some weeks ago (at which point their estimate dropped from about 100,000 to 20,000!).

What I think the app might be representative of is possibly a slight increase in infections in the young and heathy, which don't register on other things like NHS 111 or even confirmed cases as people are well and don't seek medical help. Of course that's ok as far as it goes, as long as they don't spread it to the more vulnerable.

@PatriciaHolm I only logged my 18 July test result yesterday, and the app still did not require me to specify whether it was an antigen or antibody test (though it did ask for the method by which the sample was taken, which may be a proxy for that distinction).

Dont you have to have had a positive test to be counted in the Zoe ap? So it wouldnt pick up an increase in the young who have it mildly unless or until they are tested.

One of the reasons I'm not more optimistic is that the virus could be spreading among young people who dont feel ill or do but dont test. There could be a raft of infection and it only gets picks up when they infect someone who does get tested. Test and trace are missing lots of contacts and those they do find dont always isolate.

The other reason for not feeling more confident is that while many people have been compliant up until now some will listen to an incompetent government and their social media puppets.

The Zoe app invites some people to take a test with even very mild cold/hay fever type symptoms. Then collects data on how many people have mild symptoms and uses both pieces of information together to estimate the total number of people who are positive. So it does pick up people who have only very mild illness (though not those who are completely asymptomatic). Agree it's not a problem in itself if more young healthy people are getting it mildly, but can't see how it can circulate only within that population - teenagers have parents in their 50s, 20 somethings have jobs in care homes, etc. They're not separate populations.

Does anyone have any idea why track or trace is failing to reach so many people? I'd have thought they ought to be putting some real effort in to finding out why this is? Are they missing people who are very old/sick? Who don't speak English? Due to IT issues meaning they fail to actually call some of them? Surely need to figure that out and fix it in order to improve the scheme.

Murray Strainn@murraystraining*

This week's ONS weekly death report shows there were 450 fewer deaths than the previous week
and 560 below the five year average.

I've been digging into the data to see what we can tell about the low weekly death rates this week. [Thread]

(2/6) The weeks before the pandemic hit also had fewer deaths than the previous five years.

After allowing for seasonality and scaling by mortality in the first 10 weeks of the year
we get a view of where the extra 'missing deaths' are this week.

(3/6) The bulk of the missing deaths, in absolute terms,
are at the oldest ages:
85+ for Males, 90+ for Females.

This is consistent with the some of the most frail dying a few weeks earlier due to COVID-19.

4/6) Excess deaths during the peak were seen at younger ages too (albeit at lower absolute levels)
and we're not seeing evidence of death rates there being consistently lower than expected.

(5/6) The last few weeks have also seen lower death counts than the previous five years.
The point about the missing deaths being at the oldest ages hasn't been consistently seen through this period.

(6/6) Finally, while it is good news that death rates are now lower than normal levels we have to put it into context:
excess deaths since mid-March are much higher than the dip in recent weeks.
Many loved ones have been lost too soon. #staysafe

Hancock

"on a best case scenario" we could have a vaccine by Christmas.

Presumably doesn't include the rollout of a vaccination program,
but large stocks of vaccine are being made in advance of approval

anyone analysed the lower than average deaths at the start of the year? My suspicion is that a lot of them were older too - and deaths this year were below the 5 year average at the start of the year and would have stayed that way at least until next winter. In both the old and young deaths may have returned to normal for this year - the young age group going back to having the accidents and Darwin deaths prevented by lockdown.

Apologies if this has already been discussed, I’ve been dipping in and out of this thread.

I’m concerned that we don’t really have much idea of the real scale of infection in the U.K. With the number of people tested being unknown (and before that unclear) we don’t know the percentage of positives per test done so don’t know how close we are to doing the number of tests the WHO recommends.

I’ve been keeping an eye on deaths as a better guide to infection levels assuming that deaths are estimated as 1% of infections. But the average death rate now seems to be far higher than 1% of infections 3-4 weeks ago, which suggests that we have been missing a lot of symptomatic infections to say nothing of the asymptomatic infections which could easily double the figure.

As a previous poster said perhaps this is largely mild or asymptomatic infection circulating in the younger population but means that it’s possible there is far more hidden infection out there that could potentially spread into more vulnerable populations quite rapidly given the right circumstances.

I wish I had more confidence that infections were being properly sought out by Track and Trace.

So glad I’ve found the new thread! How are the numbers looking comparatively to previous weeks today?

If older people have enough sense they will be physically keeping their distance from the young adults who are in the pub or at illegal raves. Of course you cant do that if they live in your home but part of the young population dont necessarily have a lot of contact with people old enough to become badly ill. The young living at home who infect their parents get picked up when the parents become ill and get them tested.

We are seeing higher rates of infection in communities where young people are more likely to live in multigenerational households.

@Emily - main reason for that appears to be historic reporting of deaths that occurred many weeks ago, plus over reporting by including anyone who dies having ever had a Covid death. All discussed in the previous thread. Look at the NHS figures of deaths in hospital as much more accurate (albeit not including the smaller number of deaths outside hospital)

Love these threads - been following since the first one.

On the ZOE app, I think I’ve found a significant flaw. There’s no way to update your location. As more people go on holiday to different parts of the UK, I fear the figures (especially in holiday destinations) won’t be accurate. Has anyone picked this up?

Through I saw someone tweet Tim Spector with that question recently. I’m not sure what the response was.

@alreadytaken

anyone analysed the lower than average deaths at the start of the year? My suspicion is that a lot of them were older too - and deaths this year were below the 5 year average at the start of the year and would have stayed that way at least until next winter. In both the old and young deaths may have returned to normal for this year - the young age group going back to having the accidents and Darwin deaths prevented by lockdown.

"anyone analysed the lower than average deaths at the start of the year? "

I've read several reports that this was due to flu being milder than usual

Hence presumably fewer of the very elderly dying
which is part of the "they were going to die anyway" argument

Well, yes, some of them might only have lived until the next winter flu epidemic
but the fall in total deaths in Jan+Feb 2020 looks only about 5k, much smaller than the later 65k rise that is mostly due to COVID

USA: Assigning a cause of death is never straightforward, but data on excess deaths suggest coronavirus death tolls are likely an underestimate

www.scientificamerican.com/article/how-covid-19-deaths-are-counted1/

But non-COVID conditions probably don’t explain most excess deaths, Faust said.
Only a portion of heart attack visits would have represented lives saved, he said,
because doctors must treat perhaps 10 patients to save one life.

And other causes of death - such as motor vehicle accidents - are down.

This could change with time, Faust cautioned.
For example, if cancer patients forego their treatments for a year, rather than a few months,
the impact on their death rates is much more likely to be noticeable in the population-wide data.

@alreadytaken

If older people have enough sense they will be physically keeping their distance from the young adults who are in the pub or at illegal raves. Of course you cant do that if they live in your home but part of the young population dont necessarily have a lot of contact with people old enough to become badly ill. The young living at home who infect their parents get picked up when the parents become ill and get them tested.

We are seeing higher rates of infection in communities where young people are more likely to live in multigenerational households.

After school childcare?

Several countries are finding the average infection age has fallen
e.g. the USA, Germany & other European countries
and hence the death rate is falling too

Once lockdown is lifted, many younger & some middle-aged people resume their social life and go out a lot,
whereas many of the elderly, especially the frail elderly, are still being careful and staying home much more than usual

@Firefliess

The Zoe app invites some people to take a test with even very mild cold/hay fever type symptoms. Then collects data on how many people have mild symptoms and uses both pieces of information together to estimate the total number of people who are positive. So it does pick up people who have only very mild illness (though not those who are completely asymptomatic). Agree it's not a problem in itself if more young healthy people are getting it mildly, but can't see how it can circulate only within that population - teenagers have parents in their 50s, 20 somethings have jobs in care homes, etc. They're not separate populations.

Does anyone have any idea why track or trace is failing to reach so many people? I'd have thought they ought to be putting some real effort in to finding out why this is? Are they missing people who are very old/sick? Who don't speak English? Due to IT issues meaning they fail to actually call some of them? Surely need to figure that out and fix it in order to improve the scheme.

When you are called by track and trace what number shows on your phone and do they leave a message if you don't pick up? I go through periods of not answering my phone if I'm not expecting a call and don't know the number as have little patience for scammers. I don't think that's uncommon, especially for people younger than I am. This might account for some failed tracing

Today's UK figures:

110 deaths
445 new cases

The Nobel prize banquet, held annually in Stockholm, has been cancelled ,
because of general fears about travel and continuing concerns about Sweden's infection levels

Prizes will still be awarded - cheque by post !

Yes, people not picking up would absolutely account for some failures of track and trace. I've done research with (not medical) and found that was a big problem with young people in particular. We found it was better to text them first to explain we were about to call. But big bureaucratic things like track and trace might not be able to flex to do things like that - that's just one reason why they should be working with local authorities to contact people, not replying on call centres where people have no control or autonomy to use initiative

@Littlebelina i was wondering that too, I only answer my phone if I know who it is because I hate cold callers/recorded messages. You'd hope they leave a voicemail but who knows!

@Firefliess maybe that means I'm still young it makes sense though, my parents, inlaws, grandparents would think it rude to not answer the phone

I often don't answer because I have a hearing disability, so I just use the phone to make a very occasional call

Everyone I want to hear from, knows to EMail me

Not that I deliberately don't answer the phone, just that I don't notice it and only look at it when I recharge it every few days
It's a brick too, so I can't download any App

However, the places where I actually stay for 15 minutes - gym, outdoor restaurants, hairdresser - take contact details first
So they have my address
and in Germany that means they'll use boot leather if other contact methods fail