I think we should lock down until every business in the UK fails. Then we can nationalise EVERYTHING for NOTHING and we'll be living in socialist utopia.
Yeah but they are testing way way more than they did even a month ago so there are bound to be more new cases captured, but these are likely to be people in the younger age brackets who are out socialising and whose chances of dying are miniscule.
The US states that have just had their waves saw a 37% increase average daily deaths throughout the wave and that's the peak. I can't remember what the base mortality rate was before that point but the implied IFR was around 0.31%.
The Mid West states are about to start their wave so we'll watch what happens there and hopefully learn.
It's pretty weird that people can't seem to grasp that if you have 500,000 active cases of the virus right now in the UK, the amount of tests you do doesn't change that number which is known only by God himself.
But if I randomly test 100,000 people I'm going to confirm less of active cases those than if I randomly test 200,000 people; all else equal.
And yes, we are testing more people than before, which is why the confirmed cases per 100,000 tests done charts show flat or declining numbers.
Interesting Pinkus - the data I saw I think was amalgamated for the UK or England as a whole and not regional so I can't dispute if you have a better source than I do on that. We do know hopsitalizations have dropped massively of late, coming down 50% each week from the previous which is to FeelingChills point a sign that the virus is less deadly, hitting less vulnerable people, and/or our treatments have greatly improved.
6 weeks in a row of below normal deaths suggest the virus has mainly targeted those who were going to die this year anyway. Time will tell.
6 weeks in a row of below normal deaths suggest the virus has mainly targeted those who were going to die this year anyway. Time will tell.
How do you work that out? The figures suggest that the virus has killed off some people who were going to die this year anyway, as everyone knew it would. I cannot see how the figures suggest it "mainly" killed off such people
my point is only that we have had all these discussions before (in March) and the virus is predictable - on the assumption nothing changes now (from a lockdown perspective), in a couple months we will be where we were in April/May. We can only hope new treatments can reduce the impact this time round
Guy - we know people who died from covid were mostly very old and very sick when they died. Are you disputing that? Because I think that's fairly established and it stands to reason if deaths are below average for 6 weeks running and counting, it's probably that lot that is driving down the average death count each week as they're already died. I don't think what I'm saying is a stretch or controversial. It just seems common sense, unless I've missed something. You're smart though, so feel free to point out what I've got wrong.
Without having any figures at all to back this up I believe that (in the UK and many other countries hit hard) those people most likely to get it and die have already got it and died.
We will inevitably see increases in cases with lockdown eased, people intermingling more, going back to work and so on, but I don't believe that this will result in a significant uptick in deaths. I may be wrong but I hope not.
The places where we are seeing "second waves" with a corresponding increase in deaths, most notably various states in the US, are places which never really had a first wave, either because they were just remote enough from visitors seeding infections into communities or had more stringent early lockdowns. Time will tell whether Melbourne and other parts of Australia will fall into this category or not.
As I have said before on another thread, places which were very successful early on in avoiding large numbers of cases and deaths (eg NZ) are now going to struggle to contain future outbreaks because if hardly anyone has had it then almost everyone (including those most likely to die) is still at risk.
Canadian I was not disputing the substance of what you say - covid will inevitably mean fewer deaths from other causes in the year because it will carry off Many of those who would have died anyway. The leap you made that I didn’t get was how you suggest the figures show that covid “mainly“ kills those who would die this year anyway. I am not at all sure that is true.
"The leap you made that I didn’t get was how you suggest the figures show that covid “mainly“ kills those who would die this year anyway. I am not at all sure that is true."
Well the vast majority of people that would die this year without covid-19 having been introduced are people who are very sick and/or very old, no? Outside of accidental deaths and suicides, this is what happens.
And covid-19 is killing people that are very sick and/or very old mostly, with obviously exceptions but by and large it is killing that part of the population more than any other by a wide margin. I don't think anyone debates this point, although PHE has a lot to answer for on the data side of things.
Putting that together, it stands to reason that covid *mainly* kills people who would have died of old age and/or their illness this year, but covid has sadly brought that death forward/finished them off. I don't see this as a big leap.
If that below average death count continues we may end the year with excess deaths only a few % ahead of a normal year, which would include all the (fortunately relatively few) people that died far too early from the virus. This would be as opposed to Spanish Flu, which killed millions of young and healthy people decades before their time.
I think you are making a key mistake regarding life expectancies of "sick" and "old" people. There was good piece on this on More or Less on this recently. Cant remember the exact figures given but they cited an obese man in his early 80s with a heart condition (who would be at high risk of death from COVID) would still have a normal life expectancy of 5 years. Most people killed by COVID will I think have a life expectancy of more than 1 year.
The corrollary is that the number of cases can rise without a proportionate increase in deaths because many will be younger, fitter adults who will have only a mild form...
...this is why locking down on the basis relatively small movements in the number of cases discovered is a bit weak.
Hmm Guy - fair enough, let's see how excess deaths shake out over the whole of February 20 to February 21 to get a sense of what got brought forward. lockdown prevented and caused a lot of excess deaths I'm sure which muddies it a bit. I think covid probably killed a lot more that were on the brink than those with 5 years left, but I agree it's nothing anyone can prove?
Here's one source, I got my news from a government statistician on Twitter who is really good at number crunching:
I don't think people are keeping a close enough eye on the current IFR to work out if what we are doing is proportional. As Fluffy says the number of cases per se are not important if only a tiny proportion is dying/needing hospital. This is one of the very few things I have some sympathy with Trump on.
And based on that - Guy I think you're right - no way we'll end the year with normal excess deaths. It's clearly brought a lot forward from other years.
What seems to be generally agreed is that COVID roughly doubles your chance of death in a given year (save for young kids who are far less affected). I am not good enough at maths to work out if this is relevant to the debate though...
If covid doubles your chance of death in a given year then for the vast majority of people, whose chances of death in a given year are less than 50%, covid still won't kill them. It will only kill those, statistically, with a 50% or more chance of death.
Chill, you’re just making this up as you go along and pushing out the timescales when the deaths and hospitalizations you expect don’t materialize aren’t you? It’s a long time now since ‘super saturday’
and as for guy’s ‘what seems to be generally agreed’.... m8, you’re projecting
Fluffy your chance of dying of covid if you catch it are about the same as your chances of dying this year without covid. That doesn’t relate to the figures you have given.
Proboner intuitively I think that is the wrong way of looking at it but I will confess the maths needed to link together all these stats is getting beyond my pay grade
"Fluffy your chance of dying of covid if you catch it are about the same as your chances of dying this year without covid. That doesn’t relate to the figures you have given."
I think when Prof Spiegelhalter came up with that comparison he was assuming that Covid deaths would be unabated (the early 510,000 figure with lockdown etc) and could therefore be compared to the usual 600,000 deaths a year... ...we are doing way better than 510,000 at the moment.
GSM you can see the 7 day average for new cases bottomed out on 5 July and then started to rise
this is literally the same argument everyone was having in early March. There is a delay for infections to become symptomatic and become cases (less now as more asymptomatic tests are being done) then to become hospitalisations then to become deaths
of course we don’t know the exact timeline but the pattern is inevitable
(and my main point is that talking about the death rate now is at best pointless and at worst dangerously misleading)
It's a month since 5 July now so you'd expect to have already seen the rise in hospital admissions as they come a week or two after people catch it. Wasn't early July basically when we finally got testing sorted?
Number of tests in the test and trace strategy were at their lowest level in the week ending 24 June with circa 297k tests (4355 positive). Week ending 22 July there were 366k tests (4128 positive) i.e. since week of 24 June more and more tests have been done, with 24 June week resulting in 1.4% positive tests, versus 1.1% in week ending 22 July.
Not sure what is going on, but we are getting to the point where I almost feel stats are being misused to continue a "crisis", with a view to contaminating Brexit fallout with "if it wasn't for COVID". This will then cover the volte face that might be needed in the last week of December.
0
0
I really reckon we'll get to the end of the year with deaths only up about 5% on last year which itself was a fairly low year.
0
0
I think we should lock down until every business in the UK fails. Then we can nationalise EVERYTHING for NOTHING and we'll be living in socialist utopia.
0
0
we had this with the US figures canadian - death rates will be roughly six weeks behind the epidemic
938 new cases yesterday - up from 690 on 27 July
0
0
Yeah but they are testing way way more than they did even a month ago so there are bound to be more new cases captured, but these are likely to be people in the younger age brackets who are out socialising and whose chances of dying are miniscule.
0
0
The US states that have just had their waves saw a 37% increase average daily deaths throughout the wave and that's the peak. I can't remember what the base mortality rate was before that point but the implied IFR was around 0.31%.
The Mid West states are about to start their wave so we'll watch what happens there and hopefully learn.
0
0
It's pretty weird that people can't seem to grasp that if you have 500,000 active cases of the virus right now in the UK, the amount of tests you do doesn't change that number which is known only by God himself.
But if I randomly test 100,000 people I'm going to confirm less of active cases those than if I randomly test 200,000 people; all else equal.
And yes, we are testing more people than before, which is why the confirmed cases per 100,000 tests done charts show flat or declining numbers.
0
0
I think the only meaningful guide to the number of cases is the national sample done by the ONS which I think it reports once a fortnight.
0
0
yes that was trumps argument also - and it’s true that the death % is not as high which could be improved treatment or younger people testing positive
but inevitably young people will pass it to older people or vulnerable people (who may not even know they’re vulnerable).
3 weeks after increase in cases comes increase in hospitalisations; 3 weeks after increase in hospitalisations comes increase in deaths
and its exponential unless something flattens the curve
0
0
Not in London, South East and North West. Cases are increasing and we should expect them to continue to increase over the next 6 weeks.
0
0
Chill - I'm not one for personal attacks but I really struggle to draw ascertain what conclusion you're ever attempting to reach, m8.
What point are you trying to make?
0
0
Interesting Pinkus - the data I saw I think was amalgamated for the UK or England as a whole and not regional so I can't dispute if you have a better source than I do on that. We do know hopsitalizations have dropped massively of late, coming down 50% each week from the previous which is to FeelingChills point a sign that the virus is less deadly, hitting less vulnerable people, and/or our treatments have greatly improved.
6 weeks in a row of below normal deaths suggest the virus has mainly targeted those who were going to die this year anyway. Time will tell.
0
0
6 weeks in a row of below normal deaths suggest the virus has mainly targeted those who were going to die this year anyway. Time will tell.
How do you work that out? The figures suggest that the virus has killed off some people who were going to die this year anyway, as everyone knew it would. I cannot see how the figures suggest it "mainly" killed off such people
0
0
pinkus - sorry I was replying to proboner
my point is only that we have had all these discussions before (in March) and the virus is predictable - on the assumption nothing changes now (from a lockdown perspective), in a couple months we will be where we were in April/May. We can only hope new treatments can reduce the impact this time round
0
0
Guy - we know people who died from covid were mostly very old and very sick when they died. Are you disputing that? Because I think that's fairly established and it stands to reason if deaths are below average for 6 weeks running and counting, it's probably that lot that is driving down the average death count each week as they're already died. I don't think what I'm saying is a stretch or controversial. It just seems common sense, unless I've missed something. You're smart though, so feel free to point out what I've got wrong.
0
0
Without having any figures at all to back this up I believe that (in the UK and many other countries hit hard) those people most likely to get it and die have already got it and died.
We will inevitably see increases in cases with lockdown eased, people intermingling more, going back to work and so on, but I don't believe that this will result in a significant uptick in deaths. I may be wrong but I hope not.
The places where we are seeing "second waves" with a corresponding increase in deaths, most notably various states in the US, are places which never really had a first wave, either because they were just remote enough from visitors seeding infections into communities or had more stringent early lockdowns. Time will tell whether Melbourne and other parts of Australia will fall into this category or not.
As I have said before on another thread, places which were very successful early on in avoiding large numbers of cases and deaths (eg NZ) are now going to struggle to contain future outbreaks because if hardly anyone has had it then almost everyone (including those most likely to die) is still at risk.
0
0
Canadian I was not disputing the substance of what you say - covid will inevitably mean fewer deaths from other causes in the year because it will carry off Many of those who would have died anyway. The leap you made that I didn’t get was how you suggest the figures show that covid “mainly“ kills those who would die this year anyway. I am not at all sure that is true.
0
0
Total average deaths per week is running in the 9000 - 10,000 range.
Total Covid deaths per week is running in the 200 - 300 range.
Even a large drift in the Covid number isn't going to have a large impact on the overall average...
0
0
"The leap you made that I didn’t get was how you suggest the figures show that covid “mainly“ kills those who would die this year anyway. I am not at all sure that is true."
Well the vast majority of people that would die this year without covid-19 having been introduced are people who are very sick and/or very old, no? Outside of accidental deaths and suicides, this is what happens.
And covid-19 is killing people that are very sick and/or very old mostly, with obviously exceptions but by and large it is killing that part of the population more than any other by a wide margin. I don't think anyone debates this point, although PHE has a lot to answer for on the data side of things.
Putting that together, it stands to reason that covid *mainly* kills people who would have died of old age and/or their illness this year, but covid has sadly brought that death forward/finished them off. I don't see this as a big leap.
If that below average death count continues we may end the year with excess deaths only a few % ahead of a normal year, which would include all the (fortunately relatively few) people that died far too early from the virus. This would be as opposed to Spanish Flu, which killed millions of young and healthy people decades before their time.
0
0
I think you are making a key mistake regarding life expectancies of "sick" and "old" people. There was good piece on this on More or Less on this recently. Cant remember the exact figures given but they cited an obese man in his early 80s with a heart condition (who would be at high risk of death from COVID) would still have a normal life expectancy of 5 years. Most people killed by COVID will I think have a life expectancy of more than 1 year.
0
0
where did you get the info about weekly excess deaths
0
0
The corrollary is that the number of cases can rise without a proportionate increase in deaths because many will be younger, fitter adults who will have only a mild form...
...this is why locking down on the basis relatively small movements in the number of cases discovered is a bit weak.
0
0
Hmm Guy - fair enough, let's see how excess deaths shake out over the whole of February 20 to February 21 to get a sense of what got brought forward. lockdown prevented and caused a lot of excess deaths I'm sure which muddies it a bit. I think covid probably killed a lot more that were on the brink than those with 5 years left, but I agree it's nothing anyone can prove?
Here's one source, I got my news from a government statistician on Twitter who is really good at number crunching:
https://www.expressandstar.com/news/uk-news/2020/08/04/uk-excess-deaths-during-covid-19-outbreak-down-by-1500/
0
0
I don't think people are keeping a close enough eye on the current IFR to work out if what we are doing is proportional. As Fluffy says the number of cases per se are not important if only a tiny proportion is dying/needing hospital. This is one of the very few things I have some sympathy with Trump on.
0
0
The interesting fact is that despite easing of lockdown and apparent increases in infection the number of people hospitalised hasn't increased.
0
0
And based on that - Guy I think you're right - no way we'll end the year with normal excess deaths. It's clearly brought a lot forward from other years.
0
0
Dicemand the ONS have the full figures - below is their latest summary:
0
0
What seems to be generally agreed is that COVID roughly doubles your chance of death in a given year (save for young kids who are far less affected). I am not good enough at maths to work out if this is relevant to the debate though...
0
0
If covid doubles your chance of death in a given year then for the vast majority of people, whose chances of death in a given year are less than 50%, covid still won't kill them. It will only kill those, statistically, with a 50% or more chance of death.
0
0
Guy how can it double your chance of dying if 10,000 people die on average a week and only 200 to 300 are Covid related?
0
0
Chill, you’re just making this up as you go along and pushing out the timescales when the deaths and hospitalizations you expect don’t materialize aren’t you? It’s a long time now since ‘super saturday’
and as for guy’s ‘what seems to be generally agreed’.... m8, you’re projecting
0
0
Fluffy your chance of dying of covid if you catch it are about the same as your chances of dying this year without covid. That doesn’t relate to the figures you have given.
0
0
Proboner intuitively I think that is the wrong way of looking at it but I will confess the maths needed to link together all these stats is getting beyond my pay grade
0
0
"Fluffy your chance of dying of covid if you catch it are about the same as your chances of dying this year without covid. That doesn’t relate to the figures you have given."
I think when Prof Spiegelhalter came up with that comparison he was assuming that Covid deaths would be unabated (the early 510,000 figure with lockdown etc) and could therefore be compared to the usual 600,000 deaths a year... ...we are doing way better than 510,000 at the moment.
0
0
*without lockdown
0
0
Phoebe I agree that was related to assumed IFR rate of several months ago so may need adjusting now treatment has improved
0
0
But the 500000 deaths also assumed 70 percent of the population would get it - we are nowhere near that at the moment
0
0
https://coronavirus-staging.data.gov.uk/cases
GSM you can see the 7 day average for new cases bottomed out on 5 July and then started to rise
this is literally the same argument everyone was having in early March. There is a delay for infections to become symptomatic and become cases (less now as more asymptomatic tests are being done) then to become hospitalisations then to become deaths
of course we don’t know the exact timeline but the pattern is inevitable
(and my main point is that talking about the death rate now is at best pointless and at worst dangerously misleading)
0
0
It's a month since 5 July now so you'd expect to have already seen the rise in hospital admissions as they come a week or two after people catch it. Wasn't early July basically when we finally got testing sorted?
0
0
I agree Sails, we need to be more focussed on hospital admissions to figure out how big a problem we have still got.
0
0
"GSM you can see the 7 day average for new cases bottomed out on 5 July and then started to rise"
As per this bulletin:
https://www.gov.uk/government/publications/nhs-test-and-trace-statistics-england-16-july-to-22-july-2020/weekly-nhs-test-and-trace-bulletin-england-16-july-to-22-july-2020
Number of tests in the test and trace strategy were at their lowest level in the week ending 24 June with circa 297k tests (4355 positive). Week ending 22 July there were 366k tests (4128 positive) i.e. since week of 24 June more and more tests have been done, with 24 June week resulting in 1.4% positive tests, versus 1.1% in week ending 22 July.
Not sure what is going on, but we are getting to the point where I almost feel stats are being misused to continue a "crisis", with a view to contaminating Brexit fallout with "if it wasn't for COVID". This will then cover the volte face that might be needed in the last week of December.
0
0
Sorry bananman; feelingchill won't let anyone use the excuse that more tests mean more positive cases, all else equal, because TRUMP ONCE SAID IT
Despite it being a completely logical argument as to why confirmed cases would be slightly up from one period to the next.
0
0
Yes. Chill completely debunked.
Join the discussion