B.1.1.529 Omicron variant

Covid-19 discussion, bring your own statistics
Post Reply
User avatar
Woodchopper
Light of Blast
Posts: 5147
Joined: Sat Oct 12, 2019 9:05 am

Re: B.1.1.529 Omicron variant

Post by Woodchopper » Fri Jan 07, 2022 6:37 pm

shpalman wrote:
Fri Jan 07, 2022 5:10 pm
Well it's interesting how hospital occupancy is going up, and deaths are going up (so you're hitting your target of 1000 deaths per week with quite a margin to spare, well done) but MV bed occupancy isn't.

Of course we can't see if that's because there are lots of short-stay* admissions to the MV beds, or if the doctors have realized that some patients are just going to die anyway so there's no point.


* - and of course we don't know in what state patients are leaving them.
One thing is whether or not people are dying in hospital. Could be that the deaths are at home or in care homes etc

KAJ
Fuzzable
Posts: 280
Joined: Thu Nov 14, 2019 5:05 pm
Location: UK

Re: B.1.1.529 Omicron variant

Post by KAJ » Fri Jan 07, 2022 6:39 pm

shpalman wrote:
Fri Jan 07, 2022 5:10 pm
Well it's interesting how hospital occupancy is going up, and deaths are going up (so you're hitting your target of 1000 deaths per week with quite a margin to spare, well done) but MV bed occupancy isn't.

Of course we can't see if that's because there are lots of short-stay* admissions to the MV beds, or if the doctors have realized that some patients are just going to die anyway so there's no point.


* - and of course we don't know in what state patients are leaving them.
I'm not convinced there has been a sustained increase in death rate recently.
Screenshot 2022-01-07 18.27.51.png
Screenshot 2022-01-07 18.27.51.png (114.86 KiB) Viewed 1405 times
A little while ago I read a suggestion that "Deaths within 28 days of positive test by date reported" must increase with number of positive tests, whether or not COVID causes deaths. I didn't pay much attention at the time, partly because it was (IIRC) in the Telegraph, but it seems prima facie not totally stupid. We're currently averaging well over a million cases a week. I don't know how many deaths you'd expect in that number of people who hadn't tested positive in the last 28 days.

When I get a few minutes I'll have a look at the figures for death certificate mentions, but can anyone save me the effort and point me to a discussion/rebuttal of the argument?

User avatar
jimbob
Stummy Beige
Posts: 3750
Joined: Mon Nov 11, 2019 4:04 pm
Location: High Peak/Manchester

Re: B.1.1.529 Omicron variant

Post by jimbob » Fri Jan 07, 2022 7:10 pm

KAJ wrote:
Fri Jan 07, 2022 6:39 pm
shpalman wrote:
Fri Jan 07, 2022 5:10 pm
Well it's interesting how hospital occupancy is going up, and deaths are going up (so you're hitting your target of 1000 deaths per week with quite a margin to spare, well done) but MV bed occupancy isn't.

Of course we can't see if that's because there are lots of short-stay* admissions to the MV beds, or if the doctors have realized that some patients are just going to die anyway so there's no point.


* - and of course we don't know in what state patients are leaving them.
I'm not convinced there has been a sustained increase in death rate recently.
Screenshot 2022-01-07 18.27.51.png
A little while ago I read a suggestion that "Deaths within 28 days of positive test by date reported" must increase with number of positive tests, whether or not COVID causes deaths. I didn't pay much attention at the time, partly because it was (IIRC) in the Telegraph, but it seems prima facie not totally stupid. We're currently averaging well over a million cases a week. I don't know how many deaths you'd expect in that number of people who hadn't tested positive in the last 28 days.

When I get a few minutes I'll have a look at the figures for death certificate mentions, but can anyone save me the effort and point me to a discussion/rebuttal of the argument?
It does indeed seem reasonable.

We can do some very basic sums.

https://www.ons.gov.uk/peoplepopulation ... anuary2022

1 in 15 estimated positive in England. at the moment. In 2019 England had about 500,000 deaths a year of which 48,000 were in January.

If Covid was neutral for death, then you'd expect about 3,000 of those infected now to die in January (assuming, incorrectly) that cases were spread evenly through the population rather than concentrated in the younger age groups so that's a ballpark estimate. And an overestimate.

Image
Have you considered stupidity as an explanation

User avatar
Bird on a Fire
Princess POW
Posts: 8751
Joined: Fri Oct 11, 2019 5:05 pm
Location: Portugal

Re: B.1.1.529 Omicron variant

Post by Bird on a Fire » Fri Jan 07, 2022 7:15 pm

So that's about 10 a day, for comparison with the government figures.

Has the UK ever been back down to 2019 rates of death since this pandemic kicked off?
Tree dwellers leaping out the boughs shouting «Get the paper»
Trunk hugging rebels chucking petals at a detonator

KAJ
Fuzzable
Posts: 280
Joined: Thu Nov 14, 2019 5:05 pm
Location: UK

Re: B.1.1.529 Omicron variant

Post by KAJ » Fri Jan 07, 2022 7:29 pm

jimbob wrote:
Fri Jan 07, 2022 7:10 pm
<snip>

It does indeed seem reasonable.

We can do some very basic sums.

https://www.ons.gov.uk/peoplepopulation ... anuary2022

1 in 15 estimated positive in England. at the moment. In 2019 England had about 500,000 deaths a year of which 48,000 were in January.

If Covid was neutral for death, then you'd expect about 3,000 of those infected now to die in January (assuming, incorrectly) that cases were spread evenly through the population rather than concentrated in the younger age groups so that's a ballpark estimate. And an overestimate.

<snip chart>
Bird on a Fire wrote:
Fri Jan 07, 2022 7:15 pm
So that's about 10 a day, for comparison with the government figures.

Has the UK ever been back down to 2019 rates of death since this pandemic kicked off?
Isn't 3,000 in January about 100 a day?

User avatar
Bird on a Fire
Princess POW
Posts: 8751
Joined: Fri Oct 11, 2019 5:05 pm
Location: Portugal

Re: B.1.1.529 Omicron variant

Post by Bird on a Fire » Fri Jan 07, 2022 7:32 pm

Um, yes. Sorry.

I assume the UK has been below 100 deaths a day at some point.
Tree dwellers leaping out the boughs shouting «Get the paper»
Trunk hugging rebels chucking petals at a detonator

OffTheRock
Fuzzable
Posts: 270
Joined: Mon Dec 02, 2019 6:52 pm

Re: B.1.1.529 Omicron variant

Post by OffTheRock » Fri Jan 07, 2022 7:39 pm

shpalman wrote:
Fri Jan 07, 2022 12:47 pm
... or that the same number of people are needing to go to hospital for acute reasons and when they covid-test them on arrival an increasing number of them turn out to be positive - that's probably it.

Only having the data since October means I can't really compare this to previous waves.
It’s a combination of all the things really. The BECAUSE OF Covid or WITH Covid line isn’t always very clear. For example covid is associated with an increased risk of stroke and (I think heart attacks). If you need to be admitted because you’ve had a heart attack or stroke, and have covid, that’s counted as with Covid rather than because of Covid. There might be one or two other things that applies to e.g. kidney issues.
Similarly covid can make some underlying conditions worse so some people admitted with Covid to treat an underlying condition might not have needed to be there if they didn’t have covid.
And yes, then there’s the people who were admitted and then developed covid because they were still in the incubation period when they were admitted and the people who caught it in hospital from staff, other patients or their own visitors.

It wouldn’t be a huge surprise to find the proportion of people in the WITH Covid rising given the astronomical rates in the community, which are also still rising.

And since you can’t just chuck covid positive people onto a ward with everyone else, they are still taking up beds in the covid wards, which means having to convert more wards into covid wards. It does tend to complicate their treatment for whatever it is they were admitted for in the first place.

User avatar
sTeamTraen
After Pie
Posts: 2400
Joined: Mon Nov 11, 2019 4:24 pm
Location: Palma de Mallorca, Spain

Re: B.1.1.529 Omicron variant

Post by sTeamTraen » Fri Jan 07, 2022 9:48 pm

Bird on a Fire wrote:
Fri Jan 07, 2022 7:32 pm
Um, yes. Sorry.

I assume the UK has been below 100 deaths a day at some point.
The 7-day rolling average of UK deaths was below 100 from early March until the end of July 2021, and from mid-June through early October 2020.
Something something hammer something something nail

User avatar
jimbob
Stummy Beige
Posts: 3750
Joined: Mon Nov 11, 2019 4:04 pm
Location: High Peak/Manchester

Re: B.1.1.529 Omicron variant

Post by jimbob » Sat Jan 08, 2022 11:54 am

Bird on a Fire wrote:
Fri Jan 07, 2022 7:15 pm

Has the UK ever been back down to 2019 rates of death since this pandemic kicked off?
England has in summer of 2020

ETA: the UK has in the summer of 2020 and Spring to early summer 2021

https://ourworldindata.org/grapher/exce ... untry=~GBR
Have you considered stupidity as an explanation

User avatar
shpalman
Princess POW
Posts: 6881
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond

Re: B.1.1.529 Omicron variant

Post by shpalman » Sat Jan 08, 2022 6:46 pm

molto tricky

KAJ
Fuzzable
Posts: 280
Joined: Thu Nov 14, 2019 5:05 pm
Location: UK

Re: B.1.1.529 Omicron variant

Post by KAJ » Sat Jan 08, 2022 8:48 pm

Yesterday I said
KAJ wrote:
Fri Jan 07, 2022 6:39 pm
<snip>
A little while ago I read a suggestion that "Deaths within 28 days of positive test by date reported" must increase with number of positive tests, whether or not COVID causes deaths. I didn't pay much attention at the time, partly because it was (IIRC) in the Telegraph, but it seems prima facie not totally stupid. We're currently averaging well over a million cases a week. I don't know how many deaths you'd expect in that number of people who hadn't tested positive in the last 28 days.

When I get a few minutes I'll have a look at the figures for death certificate mentions, but can anyone save me the effort and point me to a discussion/rebuttal of the argument?
jimbob helpfully responded
jimbob wrote:
Fri Jan 07, 2022 7:10 pm
<snip>
It does indeed seem reasonable.

We can do some very basic sums.

https://www.ons.gov.uk/peoplepopulation ... anuary2022

1 in 15 estimated positive in England. at the moment. In 2019 England had about 500,000 deaths a year of which 48,000 were in January.

If Covid was neutral for death, then you'd expect about 3,000 of those infected now to die in January (assuming, incorrectly) that cases were spread evenly through the population rather than concentrated in the younger age groups so that's a ballpark estimate. And an overestimate.
<snip>
Thinking about it there are many if's-and-buts in addition to those pointed out by jimbob. Essentially I'm on a hiding to nothing, but I'm starting to feel better after COVID, and it piqued me, so here are some handwaving calculations.

I need to guesstimate death rates in the absence of COVID. The ONS Deaths registered weekly in England and Wales, provisional: week ending 24 December 2021 says:
In the week ending 24 December 2021 (Week 51), 13,013 deaths were registered in England and Wales; this was 613 more deaths than the previous week (Week 50) and 12.7% above the five-year average (1,465 more deaths).
so 5 year average for week 51 is 13,013 - 1,465 = 11,548 deaths/week

I need that as a proportion of the population. ONS Population estimates gives me mid 2020 population estimates for England and Wales = 56,550,000 + 3,170,000 = 59,720,000.

That leads to an average = 11,548/7/59,720 = 0.0276 deaths/day/1,000 people.

That is only a ballpark estimate, but it gives me a way to estimate the number of deaths I'd expect from a number of people who hadn't tested positive in the last 28 days to compare with those who did.

I've calculated rolling28daysum(PublishedCases) * 11548/7/59720000 and called this "Expected"* deaths. The ratio between actual and expected published deaths is interesting**. Here's a plot on log-y scale with 7 day moving average.
image.png
image.png (49.98 KiB) Viewed 1234 times
*I know that's not a good name, but you know what I mean!
** for some value of interesting.

Herainestold
After Pie
Posts: 1798
Joined: Mon Nov 25, 2019 1:23 pm

Re: B.1.1.529 Omicron variant

Post by Herainestold » Sat Jan 08, 2022 10:07 pm

Omicron is going to prove to be deadlier than Delta.
Masking forever
Putin is a monster, Russian people are not.

User avatar
bob sterman
Catbabel
Posts: 861
Joined: Mon Nov 11, 2019 10:25 pm
Location: Location Location

Re: B.1.1.529 Omicron variant

Post by bob sterman » Sat Jan 08, 2022 10:21 pm

Herainestold wrote:
Sat Jan 08, 2022 10:07 pm
Omicron is going to prove to be deadlier than Delta.
Nice tabloid headline - but unless you're going to specify what you mean (total deaths? IFR? CFR?) what is the point in lobbing a claim like that onto a forum like this?

Herainestold
After Pie
Posts: 1798
Joined: Mon Nov 25, 2019 1:23 pm

Re: B.1.1.529 Omicron variant

Post by Herainestold » Sat Jan 08, 2022 10:49 pm

bob sterman wrote:
Sat Jan 08, 2022 10:21 pm
Herainestold wrote:
Sat Jan 08, 2022 10:07 pm
Omicron is going to prove to be deadlier than Delta.
Nice tabloid headline - but unless you're going to specify what you mean (total deaths? IFR? CFR?) what is the point in lobbing a claim like that onto a forum like this?
Probably not total deaths. It would be hard to surpass 150 000.
Deaths per period of time. More deaths in the next month/three months, than any similar time period in the pandemic.
Masking forever
Putin is a monster, Russian people are not.

Millennie Al
Dorkwood
Posts: 1334
Joined: Mon Mar 16, 2020 4:02 am

Re: B.1.1.529 Omicron variant

Post by Millennie Al » Sun Jan 09, 2022 1:11 am

Herainestold wrote:
Sat Jan 08, 2022 10:49 pm
Deaths per period of time. More deaths in the next month/three months, than any similar time period in the pandemic.
We don't (or shouldn't) care about how the deaths are spread in time. A variant that kills 1000 people on one day, but nobody on any other day is much less of concern than one that kills 10 per day for ten years. A concentration of deaths is great for headlines and making people pay attention, but not actually important by itself.

And on the subject of calculating deaths due to Covid, there is another effect which is very difficult to allow for - to the extent that it is even difficult to tell how we would count it even if we knew the numbers. That is the indirect effect of Covid due to it changing people's behaviours. For example, it has reduced commuting. How should we count the people who would have died in traffic accidents but were saved by working from home? What about people living alone and now working from home who died from something like a heart attack who would have been saved if they had been working in an office with colleagues who would have noticed and got medical assistance soon enough? These sorts of effects will show up in the total deaths, making comparison with other years include them. Yet if you try to only include people who died of Covid by looking at their medical records and death certificate, that will surely under count. It seems an extremely difficult problem to come up with much more than a crude overall figure.

Herainestold
After Pie
Posts: 1798
Joined: Mon Nov 25, 2019 1:23 pm

Re: B.1.1.529 Omicron variant

Post by Herainestold » Sun Jan 09, 2022 3:54 am

A death count of 1000 in one day is much worse because it implies a major surge of infection that would debilitate the hospital system. There would likely be a number of non covid deaths, because of problems accessing health care.

Right now it would look good if deaths decreased to 100 per day. But remember, they are mild deaths.
Masking forever
Putin is a monster, Russian people are not.

User avatar
Woodchopper
Light of Blast
Posts: 5147
Joined: Sat Oct 12, 2019 9:05 am

Re: B.1.1.529 Omicron variant

Post by Woodchopper » Sun Jan 09, 2022 8:42 am

There’s media reports going round of a ‘deltacron’ variant found in Cyprus (which looks like a combination of the two). It looks like the results were due to contamination: https://twitter.com/peacockflu/status/1 ... 93774?s=21

User avatar
shpalman
Princess POW
Posts: 6881
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond

Re: B.1.1.529 Omicron variant

Post by shpalman » Sun Jan 09, 2022 8:56 am

KAJ wrote:
Sat Jan 08, 2022 8:48 pm
Yesterday I said
KAJ wrote:
Fri Jan 07, 2022 6:39 pm
<snip>
A little while ago I read a suggestion that "Deaths within 28 days of positive test by date reported" must increase with number of positive tests, whether or not COVID causes deaths. I didn't pay much attention at the time, partly because it was (IIRC) in the Telegraph, but it seems prima facie not totally stupid. We're currently averaging well over a million cases a week. I don't know how many deaths you'd expect in that number of people who hadn't tested positive in the last 28 days.

When I get a few minutes I'll have a look at the figures for death certificate mentions, but can anyone save me the effort and point me to a discussion/rebuttal of the argument?
jimbob helpfully responded
jimbob wrote:
Fri Jan 07, 2022 7:10 pm
<snip>
It does indeed seem reasonable.

We can do some very basic sums.

https://www.ons.gov.uk/peoplepopulation ... anuary2022

1 in 15 estimated positive in England. at the moment. In 2019 England had about 500,000 deaths a year of which 48,000 were in January.

If Covid was neutral for death, then you'd expect about 3,000 of those infected now to die in January (assuming, incorrectly) that cases were spread evenly through the population rather than concentrated in the younger age groups so that's a ballpark estimate. And an overestimate.
<snip>
Thinking about it there are many if's-and-buts in addition to those pointed out by jimbob. Essentially I'm on a hiding to nothing, but I'm starting to feel better after COVID, and it piqued me, so here are some handwaving calculations.

I need to guesstimate death rates in the absence of COVID. The ONS Deaths registered weekly in England and Wales, provisional: week ending 24 December 2021 says:
In the week ending 24 December 2021 (Week 51), 13,013 deaths were registered in England and Wales; this was 613 more deaths than the previous week (Week 50) and 12.7% above the five-year average (1,465 more deaths).
so 5 year average for week 51 is 13,013 - 1,465 = 11,548 deaths/week
https://ourworldindata.org/grapher/exce ... untry=~GBR seems to be a good source for that; there's a source for deaths per week somewhere but I forget where. And the whole age-standardized thing we were arguing with Sheldrake about. But your number of 11548 deaths per week will do.
I need that as a proportion of the population. ONS Population estimates gives me mid 2020 population estimates for England and Wales = 56,550,000 + 3,170,000 = 59,720,000.

That leads to an average = 11,548/7/59,720 = 0.0276 deaths/day/1,000 people.

That is only a ballpark estimate, but it gives me a way to estimate the number of deaths I'd expect from a number of people who hadn't tested positive in the last 28 days to compare with those who did.
We can also quote that as 19.3/100,000/week to compare it to the units the UK uses for its case rates.
KAJ wrote:
Fri Jan 07, 2022 6:39 pm
... We're currently averaging well over a million cases a week. I don't know how many deaths you'd expect in that number of people who hadn't tested positive in the last 28 days.
Your calculations give 193/million/week. However, it hasn't been a million a week over all of the past 4 weeks, in fact 4 weeks ago it was half that; it's about 3 million who have tested positive over the past 4 weeks.

Whereas 1179 people have been reported over the past 7 days as dying within 28 days of a positive covid test in England and Wales. (Death certificate data is only up to 24th December at the moment so I won't use that, but it's usually even higher.)

Currently about 3 million people are being PCR tested in England in any recent 7 day period. The positivity rate used to be about 10% and now it's more like 30%.

But out of all of those 3 million you'd expect about 580 to die each week whether they'd tested positive or not. So 4 weeks ago 10% of those deaths, 58 per week, would have been attributed to covid whereas now 30% of them, 174 would be.

Whereas 4 weeks ago there were about 800 deaths per week and now there are about 1200. So no, I don't think that the extra deaths are people who would have died anyway who just happen to now have positive test results.
molto tricky

KAJ
Fuzzable
Posts: 280
Joined: Thu Nov 14, 2019 5:05 pm
Location: UK

Re: B.1.1.529 Omicron variant

Post by KAJ » Sun Jan 09, 2022 4:19 pm

@shpalman
I find your response somewhat confusing.
shpalman wrote:
Sun Jan 09, 2022 8:56 am
Your calculations give 193/million/week. However, it hasn't been a million a week over all of the past 4 weeks, in fact 4 weeks ago it was half that; it's about 3 million who have tested positive over the past 4 weeks.
I calculated a death rate which is about 193 deaths/million cases/week. Expressing it in those units doesn't mean it can only be applied when the case rate is a million/week. I had indeed said "We're currently averaging well over a million cases a week" which is true - a million a week is about 143,000 a day, the average has exceeded that every day since 28 December.
Screenshot 2022-01-09 15.13.44.png
Screenshot 2022-01-09 15.13.44.png (73.23 KiB) Viewed 1091 times
That the case rate before then was under 1 million/week doesn't mean the death rate per million/week is inapplicable - just that the calculated death rate/week will be lower than 193.
shpalman wrote:
Sun Jan 09, 2022 8:56 am
Whereas 1179 people have been reported over the past 7 days as dying within 28 days of a positive covid test in England and Wales. (Death certificate data is only up to 24th December at the moment so I won't use that, but it's usually even higher.)
I calculate the sum of deaths by date reported over the past 7 days as 1271.
Screenshot 2022-01-09 15.21.44.png
Screenshot 2022-01-09 15.21.44.png (94.95 KiB) Viewed 1091 times
shpalman wrote:
Sun Jan 09, 2022 8:56 am
Currently about 3 million people are being PCR tested in England in any recent 7 day period. The positivity rate used to be about 10% and now it's more like 30%.
I don't see a reason to restrict cases to PCR tests. My reading <clicky> is that positive LFTs are included in cases unless they are followed by a negative PCR test, i.e. positive LFTs not followed by a PCR are included. I don't see a reason to use other than the published Cases by date reported.
shpalman wrote:
Sun Jan 09, 2022 8:56 am
But out of all of those 3 million you'd expect about 580 to die each week whether they'd tested positive or not. So 4 weeks ago 10% of those deaths, 58 per week, would have been attributed to covid whereas now 30% of them, 174 would be.
So you're effectively using number of cases calculated as number of PCR tests * positivity rate. Again, I don't see a reason to use other than the published Cases by date reported.
shpalman wrote:
Sun Jan 09, 2022 8:56 am
Whereas 4 weeks ago there were about 800 deaths per week and now there are about 1200. So no, I don't think that the extra deaths are people who would have died anyway who just happen to now have positive test results.
I never suggested, or believed, that all "the extra deaths are people who would have died anyway who just happen to now have positive test results", just some of them. Using your figures, 4 weeks ago 58/800 = 7.25% "would have died anyway", now it is 174/1200 = 14.5% - a substantially greater proportion.

I've calculated values on a daily basis and plotted above the reciprocal of that percentage (PubDeathsperExp = PubDeaths/Expected rather than PercExp =100*Expected/PubDeaths) but I suggest my results are in the same order of magnitude as yours.

Code: Select all

          date PubCases PubDeaths ExpDeaths PubDeathsperExp    PercExp
1  2022-01-08   146390       313  99.14844      3.15688287   31.67682
2  2022-01-07   178250       229  96.59826      2.37064317   42.18265
3  2022-01-06   179756       231  93.28181      2.47636698   40.38174
4  2022-01-05   194747       334  89.72136      3.72263625   26.86268
5  2022-01-04   218724        49  85.75992      0.57136246  175.02025
6  2022-01-03   157758        42  80.98003      0.51864637  192.80960
7  2022-01-02   151663        73  78.04361      0.93537443  106.90906
8  2022-01-01   179637       162  75.06929      2.15800621   46.33907
9  2021-12-31   189846       203  71.29061      2.84749972   35.11853
10 2021-12-30   189213       332  67.44362      4.92263031   20.31434
11 2021-12-29   183037        57  63.70695      0.89472178  111.76659
12 2021-12-28   138831        19  59.98700      0.31673527  315.72107
13 2021-12-27   109077       143  57.24904      2.49785866   40.03429
14 2021-12-26   119923         3  55.41220      0.05413971 1847.07319
15 2021-12-25   121880        10  53.14033      0.18818099  531.40330
16 2021-12-24   122186       137  50.86650      2.69332450   37.12883
17 2021-12-23   119789       147  48.87494      3.00767630   33.24826
18 2021-12-22   109655       140  46.87084      2.98693201   33.47917
19 2021-12-21    90629       172  45.04822      3.81813071   26.19083
20 2021-12-20    91743        44  43.71826      1.00644451   99.35968
21 2021-12-19    82886        45  42.42473      1.06070212   94.27718
22 2021-12-18    90418       125  41.24015      3.03102680   32.99212
23 2021-12-17    93045       111  39.87339      2.78381147   35.92197
24 2021-12-16    88376       146  38.52525      3.78972248   26.38716
25 2021-12-15    78610       165  37.37694      4.41448657   22.65269
26 2021-12-14    59610       150  36.26239      4.13651737   24.17493
27 2021-12-13    54661        38  35.64452      1.06608255   93.80137
28 2021-12-12    48854        52  35.23137      1.47595727   67.75264
I further suggest that my plot suggests that the proportion of PubDeaths which "would have died anyway" has increased from around 1/30 a year ago to around 1/4 now.

raven
Snowbonk
Posts: 584
Joined: Mon Nov 11, 2019 8:58 pm

Re: B.1.1.529 Omicron variant

Post by raven » Sun Jan 09, 2022 10:54 pm

OffTheRock wrote:
Tue Jan 04, 2022 7:35 pm
That article says it will start to be included from the end of this month although I thought I'd read middle of Jan somewhere last month.

I know it shouldn't surprise me that it's taken omicron to speed them up on this, but it does. I dread to think how many cases we might have over the 218k reported today.
I kinda get why. Reinfections weren't so significant until Omicron, as you can see on this graph which uses a fairly crude measure of 'possible reinfection' = anyone who tests postive again after 90 days. ( It's from the latest Weekly Influenza and COVID-19 Surveillance graphs):
Reinfections UKHSA Influenza Surveillance graphs 6 January 2022.png
Reinfections UKHSA Influenza Surveillance graphs 6 January 2022.png (112.4 KiB) Viewed 1043 times
Note it uses 2 very different vertical scales. Back in the autumn when Delta was dominant first cases were running at 200,000 to 250,000 a week, reinfections were down around 3,000. So if my maths is right, maybe 1.5% of cases tops were reinfections.

Now -- and bear in mind the data for the last two weeks is incomplete and provisional -- we could be looking at ten times that number of reinfections. Of course we have more cases too, so the fraction of reinfections probably hasn't gone up quite that much, but the numbers would make a noticeable difference to the daily totals.

User avatar
lpm
Light of Blast
Posts: 4533
Joined: Mon Nov 11, 2019 1:05 pm

Re: B.1.1.529 Omicron variant

Post by lpm » Wed Jan 12, 2022 10:55 am

What the hell happened to this Omicron thing?

Why did it just fizzle? One minute it's racing up, doubling every 1.6 days at the worst. Now it's in a slump.

And hospitalisations aren't materialising like they should. The "Christmas spike" when all those infected 20 year olds met up with grandma should be appearing now but has turned into nothing at all. My guess is that the unvaccinated morons get a lot of protection from their prior infections, while the vast majority get even more protection from the vaccines, booster and prior infection.

I think what the UK now faces is another long plateau. Hospitalisations will remain high for months. Not sure what can be done about that. The plain fact of the matter is that Covid-19 is now comparable to a bad cold for the majority and a bad flu season for the vulnerable/elderly.

We've got incredibly lucky that Omi is so mild. It makes Johnson's insane recklessness look like sound judgement, which is infuriating. And nobody is going to give a sh.t about the long plateau culling 90 year olds every week.
⭐ Awarded gold star 4 November 2021

User avatar
Bird on a Fire
Princess POW
Posts: 8751
Joined: Fri Oct 11, 2019 5:05 pm
Location: Portugal

Re: B.1.1.529 Omicron variant

Post by Bird on a Fire » Wed Jan 12, 2022 11:15 am

I walked through central London yesterday evening. There's nobody there. No queues outside theatres. Empty bars. Hardly anybody in the restaurants.

Winchester has also been dead.

I suspect there's a huge amount of soft lockdown going on, on top of the usual pattern of parties in December being followed by a broke, dry January.
Tree dwellers leaping out the boughs shouting «Get the paper»
Trunk hugging rebels chucking petals at a detonator

User avatar
headshot
Dorkwood
Posts: 990
Joined: Tue Nov 12, 2019 9:40 am

Re: B.1.1.529 Omicron variant

Post by headshot » Wed Jan 12, 2022 1:28 pm

Bird on a Fire wrote:
Wed Jan 12, 2022 11:15 am
I walked through central London yesterday evening. There's nobody there. No queues outside theatres. Empty bars. Hardly anybody in the restaurants.

Winchester has also been dead.

I suspect there's a huge amount of soft lockdown going on, on top of the usual pattern of parties in December being followed by a broke, dry January.
I know of a few West End shows that have had to call it a day on their run as the audiences just aren't there to make it viable. Producers are furious at the Govt's soft lockdown, and Sunak's lack of financial aid.

Apparently pantos and other Christmas shows did ok with audiences, but I know of a few that had to close for a number of days whilst positive cases popped up and everyone went through the testing cycles.

Herainestold
After Pie
Posts: 1798
Joined: Mon Nov 25, 2019 1:23 pm

Re: B.1.1.529 Omicron variant

Post by Herainestold » Wed Jan 12, 2022 2:52 pm

Bird on a Fire wrote:
Wed Jan 12, 2022 11:15 am
I walked through central London yesterday evening. There's nobody there. No queues outside theatres. Empty bars. Hardly anybody in the restaurants.

Winchester has also been dead.

I suspect there's a huge amount of soft lockdown going on, on top of the usual pattern of parties in December being followed by a broke, dry January.

Lockdown works whether it is voluntary or not. Just wait until the caution fades, hospitalizations will explode again.
Masking forever
Putin is a monster, Russian people are not.

Herainestold
After Pie
Posts: 1798
Joined: Mon Nov 25, 2019 1:23 pm

Re: B.1.1.529 Omicron variant

Post by Herainestold » Thu Jan 13, 2022 3:20 pm

The UK has recorded its deadliest Covid-related day in 11 months.

A further 398 people were reported on Wednesday to have died with the virus.

The figure is the highest since 24 February, 2021, when 442 fatalities emerged.

Data published this afternoon also shows 129,587 new cases have been recorded over the last day.
400 per day. We are getting there.

https://metro.co.uk/2022/01/12/covid-da ... -15907643/
Masking forever
Putin is a monster, Russian people are not.

Post Reply