Summer Solstice Unlockdown
Re: Summer Solstice Unlockdown
England cases were so flat across August we can ignore lags:
England Official Cases per Day (7 day average, specimen date)
1 August 23,273
27 August 23,771
(using 27 August to avoid any bank holiday effects).
So we can simply look at hospital data /deaths at the end of August and compare to cases without worrying if it's a one week lag or three weeks or whatever. It will be within a few percent. Taking 7 day averages as at 27 August:
England admissions 769 = 3.2%
In hospital 5,890 = 24.8%
Mechanical ventilation 860 = 3.6%
England deaths 101 = 0.4%
The peak in January 2021 was:
England admissions 3,812 = 5.0x 27 August
In hospital 33,594 = 5.7x
Mechanical ventilation 3,676 = 4.3x
The previous Shpalman Equation was In Hospital = 40% Cases, which I think we can now conclude is out of date and needs to be adjusted to 25% cases.
It implies we'd need the UK case numbers to hit around 150,000 for the NHS to reach Jan-21 stress levels a couple of weeks later. My "three doublings from 30,000" estimate holds quite well, 240,000 a day would break the NHS and leave people dying in hospital car parks.
England Official Cases per Day (7 day average, specimen date)
1 August 23,273
27 August 23,771
(using 27 August to avoid any bank holiday effects).
So we can simply look at hospital data /deaths at the end of August and compare to cases without worrying if it's a one week lag or three weeks or whatever. It will be within a few percent. Taking 7 day averages as at 27 August:
England admissions 769 = 3.2%
In hospital 5,890 = 24.8%
Mechanical ventilation 860 = 3.6%
England deaths 101 = 0.4%
The peak in January 2021 was:
England admissions 3,812 = 5.0x 27 August
In hospital 33,594 = 5.7x
Mechanical ventilation 3,676 = 4.3x
The previous Shpalman Equation was In Hospital = 40% Cases, which I think we can now conclude is out of date and needs to be adjusted to 25% cases.
It implies we'd need the UK case numbers to hit around 150,000 for the NHS to reach Jan-21 stress levels a couple of weeks later. My "three doublings from 30,000" estimate holds quite well, 240,000 a day would break the NHS and leave people dying in hospital car parks.
Awarded gold star 4 November 2021
Re: Summer Solstice Unlockdown
Yes, but don't you remember all the governments and hospitals of the world desperately scrambling to try and buy more?shpalman wrote: ↑Tue Sep 07, 2021 10:01 amHadn't the idea of CPAP instead of intubation already been invented by the winter 2020-21 wave? But still, it seems like a greater proportion of the people in hospital need some kind of ventilation this time.
(The previous peak, in January this year, was about 4000 on mechanical ventilation; being a quarter of the way there would already be triggering restrictions in Italy.)
The whole Brexit's James Dyson will ride to the rescue by turning his vacuum cleaners into ventilators bollocks (only to learn that the manufacturing standards for household vs clinical are somewhat different)?
The UK had pretty much run out of ventilators first time round, so many people who needed them couldn't get 'em. This time round, they have more in stock, so even if fewer people need them, more can have them.
- shpalman
- Princess POW
- Posts: 8341
- Joined: Mon Nov 11, 2019 12:53 pm
- Location: One step beyond
- Contact:
Re: Summer Solstice Unlockdown
... but you had enough for 4000 patients last time so what does it matter now if it's 800 or 1000?
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk
@shpalman@mastodon.me.uk
- Woodchopper
- Princess POW
- Posts: 7165
- Joined: Sat Oct 12, 2019 9:05 am
Re: Summer Solstice Unlockdown
As well as thinking about the machines, you need to think about the personnel needed to provide medical care, and whether there are enough of them to cope with Covid as well as all the other reasons why people need to go to hospital.
- shpalman
- Princess POW
- Posts: 8341
- Joined: Mon Nov 11, 2019 12:53 pm
- Location: One step beyond
- Contact:
Re: Summer Solstice Unlockdown
lpm wrote: ↑Tue Sep 07, 2021 11:20 amEngland cases were so flat across August we can ignore lags:
England Official Cases per Day (7 day average, specimen date)
1 August 23,273
27 August 23,771
(using 27 August to avoid any bank holiday effects).
So we can simply look at hospital data /deaths at the end of August and compare to cases without worrying if it's a one week lag or three weeks or whatever. It will be within a few percent. Taking 7 day averages as at 27 August:
England admissions 769 = 3.2%
In hospital 5,890 = 24.8%
Mechanical ventilation 860 = 3.6%
England deaths 101 = 0.4%
The peak in January 2021 was:
England admissions 3,812 = 5.0x 27 August
In hospital 33,594 = 5.7x
Mechanical ventilation 3,676 = 4.3x
The previous Shpalman Equation was In Hospital = 40% Cases, which I think we can now conclude is out of date and needs to be adjusted to 25% cases.
It implies we'd need the UK case numbers to hit around 150,000 for the NHS to reach Jan-21 stress levels a couple of weeks later. My "three doublings from 30,000" estimate holds quite well, 240,000 a day would break the NHS and leave people dying in hospital car parks.
Who's out of date your out of date.shpalman wrote: ↑Thu Aug 19, 2021 12:13 pmNumber of Covid-19 hospital patients in England at five-month highWoodchopper wrote: ↑Wed Jun 30, 2021 7:22 am[...]
The move comes amid growing confidence that Boris Johnson will press ahead with plans to lift social distancing rules next month despite a surge in Covid cases.
He told the Cabinet yesterday that our vaccination success means Britain will be able to 'live with Covid' because the link between virus cases and hospitalisations has been broken.
Cases are still slowly rising overall so numbers in hospital are also going to keep rising which is weird considering "the link between virus cases and hospitalisations has been broken" oh wait no it hasn't.
I still have deaths with a 3 week lag and 0.3% CFR but I've put admissions on a 3-day lag at 3%. Previously I had a two-week lag but 5%.We could imagine that when things really started taking off at the beginning of June to give that peak in mid-July, those were younger people (watching the football?) with slightly lower mortality and lower risk of ending up in hospital. Yeah we could make anything up but I think the general orders of magnitude are about right and you can see that changing the lag by a lot doesn't massively change the percentage.
Numbers in hospital and on mechanical ventilation of course also depend on the rate at which patients leave hospital (which would also probably be an exponential decay) but I have those both on a three-week lag at 25% and 4% of daily new cases respectively.
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk
@shpalman@mastodon.me.uk
Re: Summer Solstice Unlockdown
Ah, OK, I had noted down your July assumptions. Your August assumptions have been pretty much proved by the outcome of the flat month. And I don't see why it would change now, given the age profile of cases is pretty much inevitable.
Awarded gold star 4 November 2021
- shpalman
- Princess POW
- Posts: 8341
- Joined: Mon Nov 11, 2019 12:53 pm
- Location: One step beyond
- Contact:
Re: Summer Solstice Unlockdown
I'm staying with 25% for hospital occupancy and 4% for MV beds, both with three-week lags.
CFR is 0.4%, three week lag.
Admissions are 3% of cases, 3 day lag.
CFR is 0.4%, three week lag.
Admissions are 3% of cases, 3 day lag.
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk
@shpalman@mastodon.me.uk
- bob sterman
- Dorkwood
- Posts: 1151
- Joined: Mon Nov 11, 2019 10:25 pm
- Location: Location Location
Re: Summer Solstice Unlockdown
That is indeed remarkably flat.
In August 2020 cases doubled. I would guess Aug 2021 has been the flattest calendar month on record.
It's certainly flat enough for the problem of lags to disappear from equations.
In August 2020 cases doubled. I would guess Aug 2021 has been the flattest calendar month on record.
It's certainly flat enough for the problem of lags to disappear from equations.
Awarded gold star 4 November 2021
- shpalman
- Princess POW
- Posts: 8341
- Joined: Mon Nov 11, 2019 12:53 pm
- Location: One step beyond
- Contact:
Re: Summer Solstice Unlockdown
It's true enough, actually: the CFR would only change from 0.4% to 0.3% if you took out the lag (which is a big relative difference but still well within the margin of error; I originally estimated 0.3% back in July anyway). It makes no difference to admissions because I only had a 3 day lag there (still 3%); hospitalizations are 20% of the case rate and MV bed occupancy 3% of the case rate instead of 25% and 4% respectively. However, these last two would stop being valid as soon as the case rate started to change substantially since they're a balance between the incoming and outgoing hospitalization rates. That means we can't really use them to predict how the hospitals would fill up if the case rate spiked due to schools and winter (unless you want to look at past data on hospitalization rates and hospital occupancy to estimate the outgoing rate constant).
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk
@shpalman@mastodon.me.uk
- shpalman
- Princess POW
- Posts: 8341
- Joined: Mon Nov 11, 2019 12:53 pm
- Location: One step beyond
- Contact:
Re: Summer Solstice Unlockdown
Actually with 156 deaths today, the average of the most recent 7 days is 140.43. But I prefer to assign this average to 3 days ago, which is the middle of the 7-day range. (The total is 983.)shpalman wrote: ↑Fri Aug 27, 2021 11:03 amWell you're already on 110 deaths per day on average, and it's going up with the same doubling time as cases have been for the past 20 days or so. It's a slow doubling time of 50 days but an average 140 deaths per day will arrive on about the 10th of September once the numbers for the 13th have come in. So the week after that, ending Sunday the 19th, will be the first one with more than 1000 deaths.Woodchopper wrote: ↑Fri Aug 27, 2021 10:08 amhttps://inews.co.uk/news/boris-johnson- ... el-1170069
A cost-benefit analysis will have set the acceptable level of Covid-19 deaths before restrictions are reintroduced at around 1,000 deaths a week, two Government advisers have told i.
Downing Street has denied it has set any “acceptable level” of Covid deaths but one adviser, who has been close to the Government since coronavirus struck 18 months ago, told i that Prime Minister Boris Johnson had privately accepted that there would be at least a further 30,000 deaths in the UK over the next year, and that the Prime Minister would “only consider imposing further restrictions if that figure looked like it could rise above 50,000”.
That's only 3 weeks from now and those deaths are inevitable because they correspond to infections which happened this week. (I previously had 0.3% CFR, but it's more like 0.4% if you ignore that peak of cases caused by young people rediscovering nightclubs and football and try to line things up with how they're going now. So you want less than about 36,000 cases per day but the current 7-day average is 34,000.)
The 7 day average might go down a bit after this because Sunday and Monday's reported numbers are always a bit low, but of course they might be a bit higher than the corresponding numbers from a week ago.
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk
@shpalman@mastodon.me.uk
- shpalman
- Princess POW
- Posts: 8341
- Joined: Mon Nov 11, 2019 12:53 pm
- Location: One step beyond
- Contact:
Re: Summer Solstice Unlockdown
They've come in, and the 7-day average centred on the 10th of September is 141.0.shpalman wrote: ↑Sat Sep 11, 2021 5:36 pmActually with 156 deaths today, the average of the most recent 7 days is 140.43. But I prefer to assign this average to 3 days ago, which is the middle of the 7-day range. (The total is 983.)shpalman wrote: ↑Fri Aug 27, 2021 11:03 amWell you're already on 110 deaths per day on average, and it's going up with the same doubling time as cases have been for the past 20 days or so. It's a slow doubling time of 50 days but an average 140 deaths per day will arrive on about the 10th of September once the numbers for the 13th have come in. So the week after that, ending Sunday the 19th, will be the first one with more than 1000 deaths.Woodchopper wrote: ↑Fri Aug 27, 2021 10:08 am
https://inews.co.uk/news/boris-johnson- ... el-1170069
That's only 3 weeks from now and those deaths are inevitable because they correspond to infections which happened this week. (I previously had 0.3% CFR, but it's more like 0.4% if you ignore that peak of cases caused by young people rediscovering nightclubs and football and try to line things up with how they're going now. So you want less than about 36,000 cases per day but the current 7-day average is 34,000.)
The 7 day average might go down a bit after this because Sunday and Monday's reported numbers are always a bit low, but of course they might be a bit higher than the corresponding numbers from a week ago.
But it does kind of look like cases had a bit of a peak about a week ago and are coming down again. Remains to be seen how much of a midweek surprise we'll get. I am starting to appreciate the lpm model of localized outbreaks amongst groups of susceptible people which quickly saturate the groups.
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk
@shpalman@mastodon.me.uk
- Woodchopper
- Princess POW
- Posts: 7165
- Joined: Sat Oct 12, 2019 9:05 am
Re: Summer Solstice Unlockdown
Two related threads.
Antibodies are have been waning among the over 70s but not in younger age groups.
https://twitter.com/john_actuary/status ... 20549?s=21
IMHO they were vaccinated first and it could also be combined with the effects of age.
Why have deaths been rising while cases have mostly been flat? The answer appears to be that deaths are occurring among the over 80s.
https://twitter.com/bristoliver/status/ ... 68482?s=21
Combined, it appears that waning immunity among older people is leading to a increasing mortality.
Antibodies are have been waning among the over 70s but not in younger age groups.
https://twitter.com/john_actuary/status ... 20549?s=21
IMHO they were vaccinated first and it could also be combined with the effects of age.
Why have deaths been rising while cases have mostly been flat? The answer appears to be that deaths are occurring among the over 80s.
https://twitter.com/bristoliver/status/ ... 68482?s=21
Combined, it appears that waning immunity among older people is leading to a increasing mortality.
Re: Summer Solstice Unlockdown
Twas ever thus, right? So are these now deaths BECAUSE of Covid, or deaths WITH Covid?Woodchopper wrote: ↑Fri Sep 17, 2021 2:13 pm[it] appears to be that deaths are occurring among the over 80s.
- shpalman
- Princess POW
- Posts: 8341
- Joined: Mon Nov 11, 2019 12:53 pm
- Location: One step beyond
- Contact:
Re: Summer Solstice Unlockdown
Woodchopper wrote: ↑Fri Sep 17, 2021 2:13 pmTwo related threads.
Antibodies are have been waning among the over 70s but not in younger age groups.
https://twitter.com/john_actuary/status ... 20549?s=21
IMHO they were vaccinated first and it could also be combined with the effects of age.
Why have deaths been rising while cases have mostly been flat? The answer appears to be that deaths are occurring among the over 80s.
https://twitter.com/bristoliver/status/ ... 68482?s=21
Combined, it appears that waning immunity among older people is leading to a increasing mortality.
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk
@shpalman@mastodon.me.uk
- Woodchopper
- Princess POW
- Posts: 7165
- Joined: Sat Oct 12, 2019 9:05 am
Re: Summer Solstice Unlockdown
You got there first.shpalman wrote: ↑Fri Sep 17, 2021 3:50 pmWoodchopper wrote: ↑Fri Sep 17, 2021 2:13 pmTwo related threads.
Antibodies are have been waning among the over 70s but not in younger age groups.
https://twitter.com/john_actuary/status ... 20549?s=21
IMHO they were vaccinated first and it could also be combined with the effects of age.
Why have deaths been rising while cases have mostly been flat? The answer appears to be that deaths are occurring among the over 80s.
https://twitter.com/bristoliver/status/ ... 68482?s=21
Combined, it appears that waning immunity among older people is leading to a increasing mortality.
- shpalman
- Princess POW
- Posts: 8341
- Joined: Mon Nov 11, 2019 12:53 pm
- Location: One step beyond
- Contact:
Re: Summer Solstice Unlockdown
Also I think "deaths been rising while cases have mostly been flat" because of the scales on the graphs he's used.
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk
@shpalman@mastodon.me.uk
- shpalman
- Princess POW
- Posts: 8341
- Joined: Mon Nov 11, 2019 12:53 pm
- Location: One step beyond
- Contact:
Re: Summer Solstice Unlockdown
The sum of the most recent seven days of reported deaths in the UK is 1003.shpalman wrote: ↑Mon Sep 13, 2021 4:30 pmThey've come in, and the 7-day average centred on the 10th of September is 141.0.shpalman wrote: ↑Sat Sep 11, 2021 5:36 pmActually with 156 deaths today, the average of the most recent 7 days is 140.43. But I prefer to assign this average to 3 days ago, which is the middle of the 7-day range. (The total is 983.)shpalman wrote: ↑Fri Aug 27, 2021 11:03 am
Well you're already on 110 deaths per day on average, and it's going up with the same doubling time as cases have been for the past 20 days or so. It's a slow doubling time of 50 days but an average 140 deaths per day will arrive on about the 10th of September once the numbers for the 13th have come in. So the week after that, ending Sunday the 19th, will be the first one with more than 1000 deaths.
That's only 3 weeks from now and those deaths are inevitable because they correspond to infections which happened this week. (I previously had 0.3% CFR, but it's more like 0.4% if you ignore that peak of cases caused by young people rediscovering nightclubs and football and try to line things up with how they're going now. So you want less than about 36,000 cases per day but the current 7-day average is 34,000.)
The 7 day average might go down a bit after this because Sunday and Monday's reported numbers are always a bit low, but of course they might be a bit higher than the corresponding numbers from a week ago.
But it does kind of look like cases had a bit of a peak about a week ago and are coming down again. Remains to be seen how much of a midweek surprise we'll get. I am starting to appreciate the lpm model of localized outbreaks amongst groups of susceptible people which quickly saturate the groups.
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk
@shpalman@mastodon.me.uk
Re: Summer Solstice Unlockdown
I'd say the same.Woodchopper wrote: ↑Fri Sep 17, 2021 2:13 pmTwo related threads.
Antibodies are have been waning among the over 70s but not in younger age groups.
https://twitter.com/john_actuary/status ... 20549?s=21
IMHO they were vaccinated first and it could also be combined with the effects of age.
I had a look at the ONS page itself though, 'cos those graphs posted in the twitter thread say 'modelled %' and I wanted to know if it was something they'd actually measured or not. Seems to come from antibody testing afaict, so I guess 'modelled' in the sense of scaled up from a sample.
Anyhow, the note for those ONS graphs says:
Linky to ONS Antibodies and Immunity blogAntibody positivity remains very high in older age groups (those aged 65 years and above) but has remained level or declined slightly since the end of July 2021. This means that some people in these groups did not have enough antibodies to be detected in the test, not that they do not have any immune protection against the virus. Please read our Antibodies and Immunity blog for more information.
tl,dr version: antibody testing doesn't look at other parts of the immune system like memory T cells, so we don't know for sure that waning antibodies means waning immunity, but that's one of the things antibody testing might shed some light on.
Re: Summer Solstice Unlockdown
Do we need waning immunity to explain that though? If vaccines don't stop you catching it, just drop you one down the severity rung as it were, then surely the hugely steep curve of risk rising with age that lpm keeps reminding us of will still mean more over 80s seriously ill and potentially dying than other age groups.Woodchopper wrote: ↑Fri Sep 17, 2021 2:13 pmWhy have deaths been rising while cases have mostly been flat? The answer appears to be that deaths are occurring among the over 80s.
https://twitter.com/bristoliver/status/ ... 68482?s=21
Combined, it appears that waning immunity among older people is leading to a increasing mortality.
So maybe the rise in deaths now is down to an initial outbreak in younger people spreading to older folk. Perhaps via people who were asymptomatic getting together with elderly relatives over the summer holidays. (I visited elderly relatives for the first time in August, and I can't be the only one. Sure, we did LFTs, but we could easily have gotten false negatives and put them at risk.)
Re: Summer Solstice Unlockdown
Ok, not much evidence for my theory in cases by age in England for the last couple of weeks:
But even though there's fewer cases in the elderly, this comparison of the 2nd peak and the last 4 weeks shows they still dominate hospital admissions:
And similarly if you look at hospitalisations per 100k population, that's highest for over 85s:
- sTeamTraen
- After Pie
- Posts: 2559
- Joined: Mon Nov 11, 2019 4:24 pm
- Location: Palma de Mallorca, Spain
Re: Summer Solstice Unlockdown
UK Covid deaths for the past 7 days were 1,009 yesterday, and are the same today (because today's number of reported deaths, 56, is the same as last Sunday's). This last happened on 14 March 2021.
Something something hammer something something nail
- shpalman
- Princess POW
- Posts: 8341
- Joined: Mon Nov 11, 2019 12:53 pm
- Location: One step beyond
- Contact:
Re: Summer Solstice Unlockdown
shpalman wrote: ↑Sat Sep 18, 2021 3:35 pmThe sum of the most recent seven days of reported deaths in the UK is 1003.shpalman wrote: ↑Mon Sep 13, 2021 4:30 pmThey've come in, and the 7-day average centred on the 10th of September is 141.0.shpalman wrote: ↑Sat Sep 11, 2021 5:36 pm
Actually with 156 deaths today, the average of the most recent 7 days is 140.43. But I prefer to assign this average to 3 days ago, which is the middle of the 7-day range. (The total is 983.)
The 7 day average might go down a bit after this because Sunday and Monday's reported numbers are always a bit low, but of course they might be a bit higher than the corresponding numbers from a week ago.
But it does kind of look like cases had a bit of a peak about a week ago and are coming down again. Remains to be seen how much of a midweek surprise we'll get. I am starting to appreciate the lpm model of localized outbreaks amongst groups of susceptible people which quickly saturate the groups.
I make it 1003 but yes, today's number of reported deaths, 56, is the same as last Sunday's, so it's still 1003.sTeamTraen wrote: ↑Sun Sep 19, 2021 3:07 pmUK Covid deaths for the past 7 days were 1,009 yesterday, and are the same today (because today's number of reported deaths, 56, is the same as last Sunday's). This last happened on 14 March 2021.
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk
@shpalman@mastodon.me.uk
-
- After Pie
- Posts: 2029
- Joined: Mon Nov 25, 2019 1:23 pm
Re: Summer Solstice Unlockdown
Thats 50 000 deaths over the year, plus whatever we see over the winter. Are we looking at 70 000 or 100 000 per year?
Masking forever
Putin is a monster.
Russian socialism will rise again
Putin is a monster.
Russian socialism will rise again
- shpalman
- Princess POW
- Posts: 8341
- Joined: Mon Nov 11, 2019 12:53 pm
- Location: One step beyond
- Contact:
Re: Summer Solstice Unlockdown
Well, it would be 50 000 deaths over the year, if the next 51 weeks were to be the same.* But I doubt they will be. Maybe the next week or two will also be similar but then it will come down, because cases are coming down quickly for some reason. Then, who knows what effects schools and autumn will have along with the growing degree of immunity in the population.Herainestold wrote: ↑Sun Sep 19, 2021 6:16 pmThats 50 000 deaths over the year, plus whatever we see over the winter. Are we looking at 70 000 or 100 000 per year?
* - there have been more than 90 000 deaths since this time last year, and more than 60 000 deaths since the beginning of 2021.
having that swing is a necessary but not sufficient condition for it meaning a thing
@shpalman@mastodon.me.uk
@shpalman@mastodon.me.uk