• Our new ticketing site is now live! Using either this or the original site (both powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!

How low would the (average) death rate for Covid have to fall for us to totally get back to normal ?

How low would the (average) death rate (per infection) have to fall for us to get back to normal ?


  • Total voters
    54
Status
Not open for further replies.

Jamiescott1

Member
Joined
22 Feb 2019
Messages
1,082
It's now up to 21%.... Assuming those voters are not taking the **** I'm deeply worried about how far from reality people's perceptions of risk are. Let's remind ourselves, that 21% are saying that if "only" 999 out of every 1000 people who are infected with Covid live, that's too risky to get rid of all these appallingly damaging anti Covid measures

I agree with you 100% but just looking at it from their perspective. If you were told that 40 people would die watching a premier league football match in the stadium (I.e 1 in 1000 people there). Would you go ?

BTW I'm all for abandoning social distancing
 
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

Bantamzen

Established Member
Joined
4 Dec 2013
Messages
9,996
Location
Baildon, West Yorkshire
I agree with you 100% but just looking at it from their perspective. If you were told that 40 people would die watching a premier league football match in the stadium (I.e 1 in 1000 people there). Would you go ?

BTW I'm all for abandoning social distancing

That's a terrible analogy, mortality is skewed very heavily towards the most elderly & those in health or care environments.
 

Class 33

Established Member
Joined
14 Aug 2009
Messages
2,362
Here is a clip of a press conference from last month where Boris says "It is my strong and sincere hope that we will be able to review the outstanding restrictions and allow a more significant return to normal from November at the earliest, possibly in time for Christmas.". Well, let's see if you deliver on what you said Boris!

See you for yourself at 12m 46s
 
Last edited:

Richard Scott

Established Member
Associate Staff
International Transport
Railtours & Preservation
Joined
13 Dec 2018
Messages
4,125
But as Bletchleyite has been pointing out recently, nothing can be 'relaxed' until the schools have
been back for several weeks and hospitalisations/deaths do not spike on a national basis.

So even as a self-confessed 'lockdown sceptic', I can see that we need to hold fire for another month or so.





MARK
How many events do we wait for? Schools did go back in June/July, what happened to cases? Nothing. We were allowed to meet up, go shopping, start using public transport, go to pubs/restaurants and what's happened? In reality nothing. Possibly in real terms cases are down as we're doing more testing now than we were we will find more positive cases. No we don't wait any longer to get moving properly. We've done enough of that.
 

Yew

Established Member
Joined
12 Mar 2011
Messages
6,880
Location
UK
It is possible for the fatality rate to change over time though, even if the virus itself hasn't - compared to March we are much better at identifying patients with Covid-19, doctors have much more experience of treating it, and there have been some advances in drug treatments. It seems quite possible, likely even, that if the same number and demographics of people infected in March and April got it now fewer would die.
It's a little macabre, but we also have fewer extremely vulnerable people too, as many will either have passed away, or have immunity. It would be an interesting study to see what percentage of those in, say, care homes, have immunity.
 

trebor79

Established Member
Joined
8 Mar 2018
Messages
4,744
It is true that the infection fatality rate is nowhere near the 12% case fatality rate, which is basically just a function of how many tests are carried out.

I take issue with claiming an IFR of 0.05% though. Using the official figures (which won't be exactly spot on, but we don't have anything better) 0.06% of the UK population has died of Covid-19. Certainly less than 100% of the population has been infected, so the IFR, whatever it is, must be greater than 0.06%. Estimates seem to vary quite a lot, but 0.5% for example might be closer. Still 25 times lower than the CFR though.

It is possible for the fatality rate to change over time though, even if the virus itself hasn't - compared to March we are much better at identifying patients with Covid-19, doctors have much more experience of treating it, and there have been some advances in drug treatments. It seems quite possible, likely even, that if the same number and demographics of people infected in March and April got it now fewer would die.
It's a little macabre, but we also have fewer extremely vulnerable people too, as many will either have passed away, or have immunity. It would be an interesting study to see what percentage of those in, say, care homes, have immunity.
I saw this interesting graph earlier today. America's second peak of infections double the first, but the peak of deaths only half the first.
Either it's getting less deadly as time goes on, or more effective treatments are being used, or some combination of the two.
Screenshot_20200828-160653.png
Graph on BBC website showing epidemic and death curves for USA.
 

Yew

Established Member
Joined
12 Mar 2011
Messages
6,880
Location
UK
There is probalby an influence of increased testing there, but It's not too bad.


Similarly, the CEBM has produced this graph, suggesting a decline in the estimated infection fatality rato.

1598638170343.png
 

talldave

Established Member
Joined
24 Jan 2013
Messages
2,424
It seems obvious, it easily kills the weakest. After that it takes time for the virus to seek out the remaining weakest (there's less of them left).
 

Bantamzen

Established Member
Joined
4 Dec 2013
Messages
9,996
Location
Baildon, West Yorkshire
There is probalby an influence of increased testing there, but It's not too bad.


Similarly, the CEBM has produced this graph, suggesting a decline in the estimated infection fatality rato.

View attachment 82854

Indeed, at the risk of agreeing with something that Trump said, the more you test, the more cases you will find. Its worth remembering though that in the early stages of the pandemic, it was generally those people entering the healthcare systems that were being tested, and thus being the most vulnerable. As testing has become more widespread, it is now finding people with the virus but not needing healthcare, or in many cases not even displaying symptoms. And so although more cases are being found, the mortality rate does not match it.
 

MattA7

Member
Joined
27 Jan 2019
Messages
473
I may have already said this but wouldn’t the case rate be a more appropriate way of estimating the risk involved with returning to normality. The death rate from the virus isn’t going to change (unless some effective treatment is found which is unlikely due to covid being a viral infection)

However if nobody gets the virus nobody will die from it.
 

Richard Scott

Established Member
Associate Staff
International Transport
Railtours & Preservation
Joined
13 Dec 2018
Messages
4,125
I may have already said this but wouldn’t the case rate be a more appropriate way of estimating the risk involved with returning to normality. The death rate from the virus isn’t going to change (unless some effective treatment is found which is unlikely due to covid being a viral infection)

However if nobody gets the virus nobody will die from it.
No as risk is mostly age dependent. If all those getting it are in their 20s/30s very unlikely many will get seriously ill, if much older then much more likely they'll be seriously ill.
If look at countries with current increasing case rates death rate doesn't seem to be following (allowing for the 21 day period between infection and death) as likely most of those are younger hence the paranoia of Governments is likely to be unfounded.
 

Dent

Established Member
Joined
4 Feb 2015
Messages
1,200
I may have already said this but wouldn’t the case rate be a more appropriate way of estimating the risk involved with returning to normality. The death rate from the virus isn’t going to change (unless some effective treatment is found which is unlikely due to covid being a viral infection)

However if nobody gets the virus nobody will die from it.
No, because:
1. The number of confirmed cases is dependent on the level of testing, which is not constant. The last month is a good example of this, where confirmed cases have increased as a result of increased testing, giving a false impression that infections were rising.
2. People with no underlying health issues getting infected but not suffering any ill effects are counted as "cases", but are not a problem.

Numbers of patients in hospital would be a much better guide.
 

Freightmaster

Verified Rep
Joined
7 Jul 2009
Messages
3,829
Numbers of patients in hospital would be a much better guide.
Like this??
EgsVNn1X0AEW1_q








MARK
 

Dent

Established Member
Joined
4 Feb 2015
Messages
1,200

DB

Guest
Joined
18 Nov 2009
Messages
5,036
The trouble we have now, with the implied-elimination strategy, is that even if the death rate is in single figures for weeks the more extreme Facebook Furloughs will be screaming 'anything above 0 is too many' and demanding that measures be retained (or even increased).
 

Richard Scott

Established Member
Associate Staff
International Transport
Railtours & Preservation
Joined
13 Dec 2018
Messages
4,125
The trouble we have now, with the implied-elimination strategy, is that even if the death rate is in single figures for weeks the more extreme Facebook Furloughs will be screaming 'anything above 0 is too many' and demanding that measures be retained (or even increased).
Government needs to start ignoring Facebook Furloughs and start getting this country back on its feet. Eventually people will start losing patience with restrictions especially when they realise how much it'll hit them in the pocket.
 

Bletchleyite

Veteran Member
Joined
20 Oct 2014
Messages
104,564
Location
"Marston Vale mafia"
1 death today despite a massive spike in cases. Is the virus genuinely evolving to be less harmful, as they often do? If so it probably is time to start reopening more things, though in reality I think we'll do schools first then think about more.
 

Richard Scott

Established Member
Associate Staff
International Transport
Railtours & Preservation
Joined
13 Dec 2018
Messages
4,125
Yes but it’s hardly a “massive spike” is it? It’s crept up a bit again. Broadly in line with expectations from the partial reopening of society.
Correct, it's around 0.003% of population of UK, not really a big deal!
 

RomeoCharlie71

Established Member
Joined
18 Sep 2017
Messages
1,788
Location
Scotland
It would be helpful if we got to know how many tests were conducted to get a test positivity rate. Not sure why PHE stopped publishing the testing data daily, as it's one of the better ways to track things. :(
 

Peter Mugridge

Veteran Member
Joined
8 Apr 2010
Messages
16,243
Location
Epsom
It would be helpful if we got to know how many tests were conducted to get a test positivity rate. Not sure why PHE stopped publishing the testing data daily, as it's one of the better ways to track things. :(


186,500 tests for the 1,715 cases announced today.
 

Yew

Established Member
Joined
12 Mar 2011
Messages
6,880
Location
UK
Unfortunately I can't remember where I read it, but I *think* some of todays cases were historical ones.
 

Dent

Established Member
Joined
4 Feb 2015
Messages
1,200

186,500 tests for the 1,715 cases announced today.

The 186,500 tests figure is actually from the 26th, it hasn't been updated since.

The Summary page shows the latest available version of each figure, but doesn't say what date each figure relates to until you go into the relevant section. Not the best designed website, and often misleading.
 

furlong

Established Member
Joined
28 Mar 2013
Messages
4,458
Location
Reading
I thought the legislation for public transport was required to be reviewed within 6 months?
.
The legislation requires continuous review and the Health Secretary is obliged to remove the restrictions as soon as they are no longer proportionate. The 6-month review they provided for is in addition to the ongoing requirement. Despite requests, no information about any such reviews has been published. Parliament (and in particular the Opposition) isn't holding him to account (as it should) and the courts remain reluctant to intervene on matters of proportionality - see Simon Dolan's appeal hearing at the end of September.

Anyone see a link with the the end of the Brexit transition period? If we end up with food, medicine and fuel shortages in January, it's very convenient for the Govt to be able to say it's illegal to gather in large numbers, to stage protest marches etc., but I guess those will happen anyway.

If you're looking for a Brexit conspiracy theory, the Dominic Cummings driving affair revealed the level of passion that remains lurking below the surface amongst those who oppose Brexit and are still hoping to derail it before the year is out. Perhaps some in the civil service and in the media have discovered Boris's Achilles' Heel and are conspiring to lead him up the garden path in his response to this crisis, trying to prise his natural supporters away from him and hoping to fatally wound his administration and salvage a closer EU relationship from it.

Regarding the poll on this thread, there's no point voting as I think the numbers are meaningless as people already discussed. Even if the virus disappears completely, testing on the current scale will still provide a lot of positive results - and the government is still talking about scaling it up by another factor of ten! If you've no symptoms and you take one of the tests and it comes back positive there's a body of opinion that reckons you've only now a 1 in 3 chance of actually having the disease - and that could easily be an over-estimate. Actual false positive rates don't appear to be known (published) but it's felt both that tests can still return positive for some time if you had the virus but already recovered (and have no live virus left) and that scaling the testing up rapidly leads to lax procedures that reduce the accuracy further. (Tests can also occasionally be positive due to material that isn't actually from this particular virus.) A testing protocol on this scale would normally require a second test independent of the first to be performed to eliminate most false positives caused by random effects before a case was counted. But statistical subtleties seem to be lost in the race to get headlines and justify past decisions.

Just to make that clearer, it's being suggested that the daily 'case' count being published may already include more people who don't have the live virus than people who do. That leads people to look at hospitalisations instead - but those numbers have their own (different) problems (namely not reporting severity of relevant and other symptoms - is this someone with low oxygen levels due to the virus or someone who had an accident and happened also to test positive). In reality you shouldn't take any one set of numbers at face value, but rather look at the totality of all the data, taking into account everything estimated about all the shortcomings, weighting it all accordingly, and then try to reconstruct possible models consistent with those weighted observations. One of the most powerful observations came early on (pre-lockdown), where it was noticed that in a significant number of households only one person showed symptoms (but in others everyone did), which suggested a sufficient level of immunity was already present in some individuals.
 
Last edited:
Status
Not open for further replies.

Top