Re: Coronavirus - Numbers and Statistics

[murraypaul]

Covid is really really small, yet N95 masks afford reasonable protection against such

So you disagree with the earlier poster who said

Magnify a human hair to the size of a old oak tree and the virus is about the size of a grain of rice next to that, masks are about as useful as trying to swat a fly with a tennis racket void of any strings.

Oh, that was you?

To quote another poster: "I'm finding your posts on this matter confusingly contradictory."

[servodude]

Covid is really really small, yet N95 masks afford reasonable protection against such

So you disagree with the earlier poster who said

Magnify a human hair to the size of a old oak tree and the virus is about the size of a grain of rice next to that, masks are about as useful as trying to swat a fly with a tennis racket void of any strings.

Oh, that was you?

To quote another poster: "I'm finding your posts on this matter confusingly contradictory."

The test of a first-rate intelligence is the ability to hold two opposed ideas in mind at the same time and still retain the ability to function
- that's not to say it isn't a test for a few other conditions also
- widdy

-sd

[zico]

Given the risks of similar small viruses that might be far deadlier, seems only reasonable that the UK should have the capacity to roll out large numbers of such masks along with other PPE independently.

That would seem reasonable, but the current government can't even manage to give proper guidance on how to wear masks effectively. Just today in the Commons, Sajid Javid appeared to be wearing a mask that didn't fit, so he had to keep fiddling with it, and the PM leads the way in badly-fitting masks.

[jfgw]

These are cases by specimen date.


I have kept the y-axis scale the same for most areas. North West, and London have different scales.
I have tried to arrange these in an "earliest peak first" sequence although this was just done by eye.



















All graphs:
My graph. Contains public sector information licensed under the Open Government Licence v3.0. : https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/


Julian F. G. W.

[1nvest]

10% of Londoners presently have Covid. Hospitals remain stressed ... due to so many staff being off in mandatory (claimed) positive test isolation. UK deaths 334 within 28 days of a positive test, a four day catch-up figure covering 1st through 4th January (83.5 daily average). On average around 600,000 people die each year in the UK (1644 simple daily average).

The vaccines appear to be inducing paranoia, perhaps anyone having had a vaccine should be prohibited from holding positions of authority.

[1nvest]

Covid is really really small, yet N95 masks afford reasonable protection against such

So you disagree with the earlier poster who said

Magnify a human hair to the size of a old oak tree and the virus is about the size of a grain of rice next to that, masks are about as useful as trying to swat a fly with a tennis racket void of any strings.

Oh, that was you?

To quote another poster: "I'm finding your posts on this matter confusingly contradictory."

The test of a first-rate intelligence is the ability to hold two opposed ideas in mind at the same time and still retain the ability to function
- that's not to say it isn't a test for a few other conditions also
- widdy

-sd

Such as the simple ability of comprehension that there are different types of masks

[XFool]

...reverse in things explaining are we.

(I love RPN!)

[servodude]

...reverse in things explaining are we.

(I love RPN!)

Stacks of fun?

[jfgw]

A comparison of the four UK countries. Bear in mind that the data are not directly comparable. Reinfections (where a positive test result follows a previously counted positive result more than six weeks previously) are counted in Wales but not elsewhere.

My graph. Contains public sector information licensed under the Open Government Licence v3.0. : https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/


My graph. Contains public sector information licensed under the Open Government Licence v3.0. : https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/


Julian F. G. W.

[jfgw]

The NHS monthly spreadsheet has been published. I have plotted the admissions by age group data.

My graph. Contains public sector information licensed under the Open Government Licence v3.0. : https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/


Julian F. G. W.

Moderator Message:
Graph replaced at poster's request (chas49)

[Julian]

Interesting to see those quite precipitous declines in the England admissions per 1000,000 in the 85 Plus and 65-84 age groups. Maybe because of the age-related prioritisation of the booster dose and many of those older people really needing it due to poorer immune response to the initial 2 doses thus even a reasonably moderate waning of immunity against severe Omicron infection leaving them with maybe only 50% protection against severe Omicron-related disease until they got the booster plus the time it took to reach full effect?

- Julian

[Lootman]

Interesting to see those quite precipitous declines in the England admissions per 1000,000 in the 85 Plus and 65-84 age groups. Maybe because of the age-related prioritisation of the booster dose and many of those older people really needing it due to poorer immune response to the initial 2 doses thus even a reasonably moderate waning of immunity against severe Omicron infection leaving them with maybe only 50% protection against severe Omicron-related disease until they got the booster plus the time it took to reach full effect?

Or maybe the weakest and most susceptible of the over-65s were killed off in one of the first few waves?

Meaning that the current population of over-65s is on average more healthy and resistant than 2 years ago?

Isn't that how herd immunity works in nature? The weakest die off but those who survive flourish. "Whatever doesn't kill you, makes you stronger" and all that.

[scotia]

A comparison of the four UK countries. Bear in mind that the data are not directly comparable. Reinfections (where a positive test result follows a previously counted positive result more than six weeks previously) are counted in Wales but not elsewhere.

Thanks for the data. I'll display the admissions (projected deaths) and reported deaths tomorrow night. Regrettably the latest English Admissions (to the 12th) seem to be back on a plateau, and the deaths by reporting date are now significantly climbing - but (hopefully) these may include a backlog of reports due to the seasonal holidays.

[jfgw]

Interesting to see those quite precipitous declines in the England admissions per 1000,000 in the 85 Plus and 65-84 age groups. Maybe because of the age-related prioritisation of the booster dose and many of those older people really needing it due to poorer immune response to the initial 2 doses thus even a reasonably moderate waning of immunity against severe Omicron infection leaving them with maybe only 50% protection against severe Omicron-related disease until they got the booster plus the time it took to reach full effect?

- Julian

ERRATUM
Ignore the graph after the peak. The rapid decline is caused my 7-day average calculation. The most recent spreadsheet is shorter (it has a later start date) and I inadvertently included some zeros in the last few days' data. I did think that the sharp peak was strange.

I have reported the post and have asked for the link to be edited.

My apologies for any confusion.


Julian F. G. W.

[9873210]

Thanks for the data. I'll display the admissions (projected deaths) and reported deaths tomorrow night. Regrettably the latest English Admissions (to the 12th) seem to be back on a plateau, and the deaths by reporting date are now significantly climbing - but (hopefully) these may include a backlog of reports due to the seasonal holidays.

I think it's far enough after the holidays that the clear climb in deaths is real. However, the climb is, as yet, relatively small compared to previous waves. So far deaths have not reached a peak anywhere, so extrapolating from prior waves is suspect; but it seems likely the total deaths may be significantly less than prior waves, if we're lucky as few as 5,000 or less than 10% of prior wave. And that would be out of many more infections than for prior waves

So far deaths in South Africa still haven't peaked. It remains possible that there will be a longer tail of delayed deaths than in prior waves.

I am also looking at New York State, which is possibly ahead of the UK, and may be a better peer to the UK than SA. NYS numbers do not look so comforting as SA. Current NYS deaths, ICU cases and hospitalizations are all trending unpleasantly. Absolute numbers are similar to Jan 2021, but cases are four times higher so on a per case basis results are better. Still there's an awful lot of sickness and dying. There are also some more pessimistic views

[scotia]

Thanks for the data. I'll display the admissions (projected deaths) and reported deaths tomorrow night. Regrettably the latest English Admissions (to the 12th) seem to be back on a plateau, and the deaths by reporting date are now significantly climbing - but (hopefully) these may include a backlog of reports due to the seasonal holidays.

I think it's far enough after the holidays that the clear climb in deaths is real. However, the climb is, as yet, relatively small compared to previous waves. So far deaths have not reached a peak anywhere, so extrapolating from prior waves is suspect; but it seems likely the total deaths may be significantly less than prior waves, if we're lucky as few as 5,000 or less than 10% of prior wave. And that would be out of many more infections than for prior waves

So far deaths in South Africa still haven't peaked. It remains possible that there will be a longer tail of delayed deaths than in prior waves.

I am also looking at New York State, which is possibly ahead of the UK, and may be a better peer to the UK than SA. NYS numbers do not look so comforting as SA. Current NYS deaths, ICU cases and hospitalizations are all trending unpleasantly. Absolute numbers are similar to Jan 2021, but cases are four times higher so on a per case basis results are better. Still there's an awful lot of sickness and dying. There are also some more pessimistic views

Thanks for the info. Yes - I was trying to be a bit optimistic on the reported deaths, but I fear that they are climbing. However, as you say, it doesn't look like any way as bad as the pre-vaccine peaks.

[scotia]

My apologies for any confusion.
Julian F. G. W.

Thanks for the update - the reporting dates have been erratic over the seasonal holiday period.

[Julian]

Thanks for the data. I'll display the admissions (projected deaths) and reported deaths tomorrow night. Regrettably the latest English Admissions (to the 12th) seem to be back on a plateau, and the deaths by reporting date are now significantly climbing - but (hopefully) these may include a backlog of reports due to the seasonal holidays.

I think it's far enough after the holidays that the clear climb in deaths is real. However, the climb is, as yet, relatively small compared to previous waves. So far deaths have not reached a peak anywhere, so extrapolating from prior waves is suspect; but it seems likely the total deaths may be significantly less than prior waves, if we're lucky as few as 5,000 or less than 10% of prior wave. And that would be out of many more infections than for prior waves

So far deaths in South Africa still haven't peaked. It remains possible that there will be a longer tail of delayed deaths than in prior waves.

I am also looking at New York State, which is possibly ahead of the UK, and may be a better peer to the UK than SA. NYS numbers do not look so comforting as SA. Current NYS deaths, ICU cases and hospitalizations are all trending unpleasantly. Absolute numbers are similar to Jan 2021, but cases are four times higher so on a per case basis results are better. Still there's an awful lot of sickness and dying. There are also some more pessimistic views

Thanks for the info. Yes - I was trying to be a bit optimistic on the reported deaths, but I fear that they are climbing. However, as you say, it doesn't look like any way as bad as the pre-vaccine peaks.

I’m becoming increasing sceptical of the deaths data assuming that we’re talking about people dying within 28 days of first testing positive for Covid-19 (henceforth “28DA”). I think that at the moment one really needs to look at the mentions-Covid-19-on-death-certificate numbers (henceforth “CODC”) and just accept the lag on that data.

My concern with the 28DA data is due to the massive amount of testing that we are doing now and the huge number of positive cases each day. Throughout the pandemic, or at least since testing on any reasonable scale started, it has been the case the some percentage of all the people dying on any given day from heart attacks, strokes, accidents etc some percentage will have tested positive for the first time in the previous 28 days but now with so many more tests being done, we have already hit over 2,000,000 tests a day on 2 days in January, and the case numbers showing that there is still a fairly high positivity rate that must surely mean that the percentage of those not-Covid-19 deaths that get dragged into the 28DA figures increases.

I know there is the counter effect of a genuine Covid-19 death not making it into the 28DA figures because that person had already tested positive over 28 days before death but with Omicron now looking more and more like it is less severe hence very possibly causing fewer genuine deaths I think the accidental overcounting effect is probably swamping the accidental undercounting effect.

Neither of those distortions happen with the CODC numbers but there is a delay in compiling that data due to the time it takes for the death certificates to be processed so I can see why the 28DA data is still published but I do worry that with the massive amount of testing and probably quite a few people testing themselves for the first time in the run up and over Christmas the distortions on the 28DA data might be in danger of making those data a very unreliable indicator now.

- Julian

[servodude]

Neither of those distortions happen with the CODC numbers but there is a delay in compiling that data due to the time it takes for the death certificates to be processed so I can see why the 28DA data is still published but I do worry that with the massive amount of testing and probably quite a few people testing themselves for the first time in the run up and over Christmas the distortions on the 28DA data might be in danger of making those data a very unreliable indicator now.

Indeed

it was useful for "dead reckoning" during the fog at the start of the pandemic when things were moving faster than the paperwork could keep up
- it has served its purpose and should be retired

the pun was accidental
-sd

[9873210]

Neither of those distortions happen with the CODC numbers but there is a delay in compiling that data due to the time it takes for the death certificates to be processed so I can see why the 28DA data is still published but I do worry that with the massive amount of testing and probably quite a few people testing themselves for the first time in the run up and over Christmas the distortions on the 28DA data might be in danger of making those data a very unreliable indicator now.

Indeed

it was useful for "dead reckoning" during the fog at the start of the pandemic when things were moving faster than the paperwork could keep up
- it has served its purpose and should be retired

the pun was accidental
-sd

The two numbers track very closely. Between Jan 1 2021 and Dec 15 2021 the r-square between the seven day averages of 28DA and the CODC is 0.99!!!! (and the death certificate numbers are 7% higher).

I think we can reasonably use 28DA as a timely proxy for CODC. In any case the alternatives that don't use it are worse. Either you make policy without data or you make policy based on data such as cases, which would result in a massive overreaction to Omicron.



!!!! I rarely use multiple explanation marks, but I have never seen an R^2 of .99 in real data. I was ready to work up a multi-variant model that included test numbers. But with only 1% left to explain it scarcely seems worth it.