Daily numbers, graphs, analysis thread 14

[BigChocFrenzy]

Welcome to thread 14 of the daily updates

Resource links:

Uk dashboard deaths, cases, hospitals, tests - 4 nations, LAs, English regions
Slides & data UK govt pressers
[[https://www.gov.uk/guidance/coronavi
rus-covid-19-information-for-the-public UK stats]] list of reports added daily by PHE & DHSC
PHE Surveillance report infections & deaths released every Thursday with sep. infographic
ONS England infection surveillance report ONS UK statistics for CV related deaths, released weekly each Tuesday
Daily ECDC report UK & EEA
Worldometer UK page
Plot FT graphs compare countries deaths, cases, raw / million pop
Covidly.com world summary & graphs
Plot COVID Graphs Our World in Data additional data

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

news.sky.com/story/coronavirus-uk-death-toll-falls-by-5-377-as-government-changes-way-fatalities-are-counted-12047827

@Piggywaspushed

Well done BCF. Was it removed?

... Yes it was removed MN will delete significantly false / misleading info or conspiracy nuts

As the site was notorious enough to be on Wiki, I reported with the WIki link

Reports for weeks have been of a big fight over changing the data:
epidemiologists vs politicians / those with a political eye

"Epidemiologists say this 60-day measure is more accurate and will urge officials to use this
The government's own analysis of data in England found 96% of deaths occurred within 60 days or had COVID-19 on the death certificate. 88% of deaths occurred within 28 days.

Even if politically the lower figure must be used publicly, I assume epidemiologists will still know and use the other figure for their calculations, if they think it more accurate.

However, it seems only logical that the Uk should use the same system as other countries for public statements,
even if other figures are used by the professionals

ONS states that 56,800 deaths registered in the UK with Covid-19 on the death certificate.

So the discrepancy to ONS will increase

Think coronavirus.data.gov.uk/ still shows old deaths data at the moment, or am I missing something?

@ancientgran significant numbers of people whose normal place of residence was a care home were dying in hospital of Covid-19 at the peak of the pandemic. www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/articles/deathsinvolvingcovid19inthecaresectorenglandandwales/deathsoccurringupto1may2020andregisteredupto9may2020provisional

Therefore it is unlikely that many infected patients were sent to care homes. Care homes cant be forced to take people so clearly the home believed they could manage the infected person. No I wouldnt have sent them out - but if I was a care home owner I wouldnt have accepted them either if I couldnt keep the rest of the residents clear of infection. Some homes did manage to keep their residents free from infection.

The Nightingales were not available at the start of the pandemic when patients were being sent out. When one did become available it was a political dick waving exercise, there were never people to staff it. With no tests for staff or their families and limited PPE there were less bodies around - and few of them properly trained. As for these empty beds you think are lying around - the NHS runs close to capacity in winter, if beds are empty it's usually because they cant staff them.

It would be useful if the data on deaths was clearer, the detail doesn't need to be lost for that because there are many categories that would be useful for different purposes:

1. International standard 28 days, for those countries that track it properly, for comparison. This should include or exclude care home deaths dependent on what that standard is, or copy Germany if there isn't one.
2. Additional England-tracked deaths: 60 + care home if directly attributable and not already included in 1 + special long tail cases that are nothing else = the number of "real" deaths
3. Additional other deaths with primary cause non-covid, where covid was mentioned as present (some may be aggravated and many others not, wiser to give it its own category)
4. Lockdown-induced excess deaths: excess missed treatment
5. Recession-induced excess deaths
6. Excess flu deaths (some during summer, there will be more in winter)
7. Other excess deaths (if the number goes over 1000 then someone should figure out common causes / missed assignments).

Really like your charts and tables @boys3 Especially the decile one. But can I ask you split the LAs into deciles? Are they split differently week by week? Or are you tracking the change in a fixed group of LAs based on their rates at the start/end/overall? Ie - does the composition of each decile change week by week?

@Firefliess - yes, to your last question, and I think that it the best approach. But tracking LAs and their movement is something I should get on and do. Plus the current decile breakdown in for the 149 UTLAs, it might actually be better to apply it to the full list of LAs removing counties and adding all the districts in.

I'd best get cracking :)

Agree on greater clarity on death stats

28 days is stark cut off given the lag and without the 28 days is even stranger

Have just sent this to a friend I was speaking to who had no idea about the symptoms other than the NHS's "main" ones: others may wish to use it to educate people!

www.who.int/health-topics/coronavirus#tab=tab_3

Most common symptoms:
fever.
dry cough.
tiredness.

Less common symptoms:
aches and pains.
sore throat.
diarrhoea.
conjunctivitis.
headache.
loss of taste or smell.
a rash on skin, or discolouration of fingers or toes.
_
Serious symptoms_:
difficulty breathing or shortness of breath.
chest pain or pressure.
loss of speech or movement.

(Not sure even I'd hear of the very last one, would seem more like a stroke. Interesting that they haven't got anosmia as a main symptom too.)

Germany and Ireland, Belgium, France, Sweden all include care home deaths and deaths in other institutions - prisons, homes for children and the severely disabled, refugee centres etc

I see no justification for excluding any of these

Good, they can be in category 1 then

I completely agree that hospitals or any other additional buildings or equipment are no use without sufficient trained staff to use them

That UK govt report I posted upthread looked the first attempt to quantify deaths - and lost years - from 3) - 7)\

That is something which takes weeks to do, so can't be in the daily figures, just in reports every couple of month or so.

imo, the daily figures need to be
a) with the 28 cutoff and
b) with the 60 days
plus updating a) as some of the deaths in b) are shunted over after investigation

Do the UK still include deaths outside of hospital?

I know initially they did not. I hope that doesn't change.

As mentioned here many times, excess deaths are the best indicator and hopefully over time we will get a better idea of how many are indirectly as a result of Covid.

And it is yet to be seen how excess deaths will be over the remainder of the year and indeed next year will be.

I'm not sure my comment was clear. I hope we continue to record all deaths no matter the setting.

Therefore it is unlikely that many infected patients were sent to care homes.

Unfortunately, based on the SAGE reports, looks like this is exactly what triggered the pandemic in care homes.

I have previously given a link to the genome analysis of London care home transmissions:

www.cogconsortium.uk/wp-content/uploads/2020/07/REPORT-ON-USE-OF-SARS-CoV-2-GENOMICS-TO-UNDERSTAND-TRANSMISSION-28-June-2020.pdf

CARE HOMES

One of the first studies to be completed in the UK by Shamez Ladhani, Maria Zambon and others describes an investigation in six London care homes reporting suspected COVID-19 outbreaks during April 2020. Residents and staff had nasal swabs taken for SARS CoV-2 testing using RT-PCR and were followed-up for 14 days.
They found that 107/268 (39.9%) residents and 51/250 (20.4%) staff were SARS CoV-2 positive. The 158 PCR positive samples were sequenced and 99 (68 residents, 31 staff) distributed across all the care homes yielded sufficient sequence data for analysis. This identified multiple independent introductions into each care home, rather than a single introduction followed by within-care home transmission. Several introduction events were followed by considerable within-care home transmission, although there were numerous instances where an introduction event was not followed by any detected forward transmission. Clusters commonly contained isolates from both residents and staff members, although it is not possible to infer directionality. There was no instance where clusters contained isolates from different care homes, refuting the suggestion that staff working in multiple care homes had transmitted SARS CoV-2 between different homes in this investigation.

In summary:

• There is no evidence that COVID was transmitted by temporary staff moving across care homes
• However, there is evidence that COVID was introduced into London care homes multiple times at approximately the same time. The timing of transmission coincided with hospitals discharging care home residents to create capacity for COVID patients.

I am guessing that the vast majority of the discharged residents did not have COVID symptoms and might have been feeling absolutely fine after having recovered from a different procedure. However, they have not been tested and it looks highly likely they were the source of initial transmission bringing COVID from hospitals into the care home environment.

Yes, so far at least there has been no change re settings

The discrepancy with ONS - who provide the excess deaths data - seems due to having more information after deaths are registered

There were complaints back from some care homes that they had been forced to take infected patients under possible threat of in the future not receiving residents paid by local authorities

@boys Gotta put those statistical skills of yours to good use and keep you busy! I think you're right that you're kind of double counting some places if you have both counties and districts in your list. I would take the counties out (unless they are unitaries, eg Cornwall) so you have just lower tier authorities and unitaries in there. Then everywhere is in your list once and once only.

I do find that analysis by decile useful though in understanding whether rates are increasing everywhere, or just in certain locations where we know there is a focus on a lot more testing.

On a positive note, hospital admissions are still falling gradually. I find it hard to believe the increase in cases could really be indicative of an actual increase in Covid among the young, and them still not passing it on to older people.

Sorry, first use of my new Apple keyboard, allegedly "Smart"

www.theguardian.com/world/live/2020/aug/12/coronavirus-live-news-lebanon-sees-record-cases-after-blast-as-new-zealands-biggest-city-heads-into-lockdown?page=with:block-5f342b118f085d9bdb7e68d2#block-5f342b118f085d9bdb7e68d2

France reported 2,524 new coronavirus infections over the past 24 hours, a new post-lockdown daily record,
but there was no strain on hospitals as the virus circulates mainly among younger people,
the health minister said.

[quote Firefliess]@boys Gotta put those statistical skills of yours to good use and keep you busy! I think you're right that you're kind of double counting some places if you have both counties and districts in your list. I would take the counties out (unless they are unitaries, eg Cornwall) so you have just lower tier authorities and unitaries in there. Then everywhere is in your list once and once only.

I do find that analysis by decile useful though in understanding whether rates are increasing everywhere, or just in certain locations where we know there is a focus on a lot more testing.

On a positive note, hospital admissions are still falling gradually. I find it hard to believe the increase in cases could really be indicative of an actual increase in Covid among the young, and them still not passing it on to older people. [/quote]
No double counting, rather I'd been using the shorter LA list to date - eg just the genuine top tier authorities - unitaries, mets and two tier area county councils. Have now moved to all LAs with the two tier area county councils removed.

That has yielded a raising of eyebrows - mainly before P2 cases started being included on the dashboard - as the overall figures in county areas can mask some extremes in their constituent districts.

South Holland (lincs) - highest weekly cases per 100,000 for three consecutive weeks back in May, with one of those at close to 170 cases per 100,000 in a single week (so higher than any Leicester weekly cases per 100,000 figure).

It takes time to filter through, it did before. We also don't know what percentage are asymptomatic of the ones being found, sadly I think not many but maybe picking up milder cases who would have thought it couldn't be covid had it been back in March.