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.

Is this research all wrong too?

Greater risk of severe COVID-19 in non-White ethnicities is not explained by cardiometabolic, socioeconomic, or behavioural factors, or by 25(OH)-vitamin D status: study of 1,326 cases from the UK Biobank

Not sure what you mean.

It says right there, it's not explained by those factors. It doesn't mean it's not explained by other factors.

I have seen basically three angles on this:

1. the standard scientific response, which is to study the data and adjust for different variables, but not to assert 'COVID-19 is racist'
2. the slightly more sociological response, which is to also study the data but to have a pre-determined conclusion to overlay onto the data
3. the politician or journalist who hasn't read the study, but wants a simple solution 'COVID-19 racist'

I've seen cut offs at 65+, 70+, and 80+ all mentioned, but regardless, I think an important consideration is that a vast number of that group are not elderly frail

Not true. The reason people are dying in greater numbers in these age groups is precisely because they ARE frail in some way or another. Granted, you can have very fit 65+ and 70+, all my brothers fit into this category, with no health issues, or medication. However, once you get to 80+ your immunity sucks, you are likely to have one or two or three major health issues, cancer, heart, stroke, diabetes, mobility, dementia, alzheimers, and taking several different types of drugs a day.

This is a happy read, and gives us all something to look forward to in later life.....

www.ageuk.org.uk/Documents/EN-GB/For-professionals/Research/Age_UK_almanac_FINAL_9Oct15.pdf?dtrk=true

"the proportion of patients with one or more diagnosed
disease(s) increases from the youngest groups onwards, plateauing at over 90% in the 85-89, 90-94 and 95-99 groups"

Just watching todays briefing.

A further 359 people have died with covid, which to me feels quite high for a Wednesday. Isn't the trend that it's higher on Tuesday because of the weekend lag?

It looks like it's plateauing to me.

The reason people are dying in greater numbers in these age groups is precisely because they ARE frail in some way or another. Granted, you can have very fit 65+ and 70+, all my brothers fit into this category, with no health issues, or medication. However, once you get to 80+ your immunity sucks, you are likely to have one or two or three major health issues, cancer, heart, stroke, diabetes, mobility, dementia, alzheimers, and taking several different types of drugs a day.

Well I think that no matter how fit they say they are 65 yo are in fact far more likely to drop dead than 25 yo. There's a phenomenon of retired people who cycle thousands of miles a year and are ostensibly very fit dropping dead on rides. Like it or not, if you are 65, your body is all kinds of used up in ways that a 35 year old's is not.

359 definately disappointing.

@sanitydecreasing

Does this help? This is England, all settings; Orange is announced number, black is by day of actual death. For example, 40% of hospital deaths announced today were from more than a week ago.

The reporting cycles for the various inputs to the overall total are all over the place. We do get a weekend lag, but PHE (who submit the care home data) are all over the place regularly. The day by day trend of death by actual day of death is down.

The graph is courtesy of a very good Twitter account, @RP131, who analyses the data every day.

once you get past 65 virtually everyone is on at least one medication. That does not, however, mean they are "frail", their health problems may be well controlled and they may be leading a more active life that when they were working. The voluntary sector largely runs on the newly retired. These supposedly "frail" people are often providing childcare for grandchildren too.

The biobank study had to use some data from when people enrolled in the study. That was 10-14 years ago. Some measures may not change in that time but I would like to see a study on vitamin D that used recent data. Also not clear how they defined the variable for that - continuous or, as the figure suggests, putting it into 3 bands. They also seasonally adjusted the data, presumably to estimate the level at time of infection.

That looks much better PatriciaHolm.

Although its been explained over and over again that a seven day rolling average is the best way to look at trends, not single days, it's still difficult to try not to extrapolate out of each day's figures.

Although, if some of today's announced deaths were a week or more ago, that implies that the black line for the latest week or so is going to go up when more deaths are announced over the next week. The plateau could well still be happening.

Todays figures included quite a lot of backdated deaths, some in March!

looking at the deaths per day in number form(Hospital only) rather than graph it does show it's consistantly going in the right direction but seems to be plateauing and will take a while to get down to very low numbers, on the cusp of the first less than hundred day soon.

01-May 305
02-May 266
03-May 251
04-May 257
05-May 249
06-May 261
07-May 254
08-May 212
09-May 201
10-May 193
11-May 163
12-May 183
13-May 159
14-May 176
15-May 169
16-May 167
17-May 136
18-May 150
19-May 140
20-May 150
21-May 144
22-May 120
23-May 120
24-May 110
25-May 129
26-May 131
27-May 113
28-May 111
29-May 98
30-May 80
31-May 65
01-Jun 66
02-Jun 20

Do you think there’s a second wave coming?

Thank you for sharing that graph and Twitter account @PatriciaHolm. I knew not to take each day's reported figure as black and white but still felt worried seeing it was so close to yesterday's (Tuesday's generally being high).

Also it looks like (although I suspect the effect is small) from the 1st June PHE started including deaths where there was a positive test from pillar 2 (processed in a commercial lab). It means previously we weren't counting these deaths but they added historic ones from these tests as a lump on Monday (those additional 450 deaths which were commented on on the 1st).

Second wave - if coming will take several weeks for cases to pick up (Longer than before as people not gathering in packed crowds) and another week or so on top of that for hospital admissions to increase. We just have to wait and see. I suspect a second peak mid July when schools are due to close again anyway might be on the cards.

The thing is that they noticed that the holidays helped temper the spread of swine flu.,

Unfortunately, it is not surprising that hospital deaths are plateauing. Based on the latest INARC report, we have still had 1,285 patients in ICU. Based on the trends so far, nearly half of them, sadly, may not survive.

Thank you @PatriciaHolm

Also, I noticed that hospital admissions have plateaued over the last 3 weeks, a few spikes here and there. We many find we rattle on at around this level for a while?

@NeurotrashWarrior

Also, I noticed that hospital admissions have plateaued over the last 3 weeks, a few spikes here and there. We many find we rattle on at around this level for a while?

That would make sense I suppose, if r0 is only just under 1 and lockdown has been in place so long that all cases have been transmitted under lockdown conditions but people mixing/shopping more than in the first month or two of it. I can see it trundling along like this for a few more weeks at least. Flare ups in healthcare and potentially a few schools. I'm still hopeful figures will be dramatically lower by the end of the month though.

maggieFS Yes, the FT tool is not normalised wrt population

Also, I noticed that hospital admissions have plateaued over the last 3 weeks, a few spikes here and there.

I have been checking hospital admissions because I was also wondering about the perceived spike last week but it looks like the numbers are reported on the same basis as deaths and test results (i.e. based off cumulative totals). As a result, there is always a drop over the weekend and spikes on Mon and Tue.

The overall trend is still down (these are England only numbers from daily government briefings):

12/5 - 804
13/5 - 788
14/5 - 736
15/5 - 718
16/5 - 678
17/5 - 639
18/5 - 637
19/5 - 697
20/5 - 713
21/5 - 675
22/5 - 685
23/5 - 595
24/5 - 471
25/5 - 472
26/5 - 475
27/5 - 552
28/5 - 562
29/5 - 545
30/5 - 479
31/5 - 436

Christian Drosten (chief virologist and advisor to the German government) has said we may well avoid a 2nd wave here.

So the Uk could also do so, if lockdown is relaxed sensibly in stages.

The key learning they have been discussing here is that most infected people infect only a few others or even none,
but a minority infect a very large number of people,
Hence, banning large crowds is vital, as is large-scale contact tracing for every new outbreak.

Drosten and Streeck (Gangelt study) say we should continue to ban potential superpreader events e.g. carnivals, large sports & shows
and lockdown locally whenever new outbreaks flare up as they have already done in a few care homes, meat plants etc

Here the fatality rate in ICU has been a bit under 30%,
but those who remain may have been there for weeks or be the most ill - no data on that yet - so we can't assume that 70% of the remaining ICU patients will survive

Similarly in the UK, the last remaining ICU patients may have a higher casualty rate,
which could distort the statistics,
as these would be people who became infected some time ago.

Kai Kupferschmidt

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

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.
.....
That’s why in addition to R, scientists use a value called the dispersion factor (k), which describes how much a disease clusters.
The lower k is, the more transmission comes from a small number of people
....
In the flu pandemic of 1918....the value was about one, indicating that clusters played less of a role.
^.....
Estimates of k for SARS-CoV-2 vary.
.....
in a recent preprint, Adam Kucharski of LSHTM estimated that k for COVID-19 is as low as 0.1.
“Probably about 10% of cases lead to 80% of the spread,” Kucharski says.

@BigChocFrenzy

Interesting article. I wouldn’t be surprised the mode and median number that an infected person infects is zero...