Number of Deaths

[Harry Wilson's tackle]

Here is my update. If the fall in increase in deaths is catching up with the fall in the increase of cases, this would 'fit' with the overall idea that everywhere (even in Brazil if the latest very sharp fall is correct) is on that part of the new cases curve where the numbers are falling very slowly as shown in fig 6 of the data posted by [MENTION=36]Titanic[/MENTION] yesterday.





Blimey, it's slow in the UK, though.

 

[WATFORD zero]

Thought this was an interesting addition to the stats thread as there are stats specifically to compare excess deaths per country. Probably the most effective comparison basis that I have seen.

Need to note that the date of the latest figures is different per country, but will hopefully be updated going forward.

Excellent coverage on this from the FT. Not pretty reading for the UK though.


https://twitter.com/jburnmurdoch/status/1260233393838620675?s=21

 

[dazzer6666]

Thought this was an interesting addition to the stats thread as there are stats specifically to compare excess deaths per country. Probably the most effective comparison basis that I have seen.

Need to note that the date of the latest figures is different per country, but will hopefully be updated going forward.

Agree it's the best comparison as most other datasets are so flawed or have different bases. Interesting the us, Spain, Italy, Belgium and Netherlands are very similar (so far). There will be dozens of theories around this - going to leave it to the geeks to properly analyse...................

 

[Harry Wilson's tackle]

As promied, I had a chat with my modeller mate today. This is what he said when I asked him what he thought....

"The government has made a hash of this across almost every index. I think that the fall out could be fairly large when people are told the this government's failures led to their loosing their grandparents and that in other countries that followed the science they had only a fraction of our deaths."

Stark.

 

[Weststander]

Agree it's the best comparison as most other datasets are so flawed or have different bases. Interesting the us, Spain, Italy, Belgium and Netherlands are very similar (so far). There will be dozens of theories around this - going to leave it to the geeks to properly analyse...................

The ‘good’ thing about total excess death stats is that an agreed formula is used across the globe, and it takes away most of the debates. The only proviso, can we actually trust the total deaths figures per se from the likes of China, Russia and Iran? National pride intervening. If their total excess death stats, per million, look suspiciously low, they might be disregarded.

 

[dazzer6666]

The ‘good’ thing about total excess death stats is that an agreed formula is used across the globe, and it takes away most of the debates. The only proviso, can we actually trust the total deaths figures per se from the likes of China, Russia and Iran? National pride intervening. If their total excess death stats, per million, look suspiciously low, they might be disregarded.

Frankly I’d ignore China, Russia and Iran for obvious reasons. Let’s see how we compare with elsewhere - rest of Europe primarily.

 

[Weststander]

Frankly I’d ignore China, Russia and Iran for obvious reasons. Let’s see how we compare with elsewhere - rest of Europe primarily.

That’s my intention.

The USA is a difficult one to compare to European nations too. Effectively half a continent, with the megalopolis of NYC/NJ suffering like Paris, London and Madrid, whilst huge swathes in the centre are seemingly little scathed. Better imo to look at individual states.

 

[beorhthelm]

As promied, I had a chat with my modeller mate today. This is what he said when I asked him what he thought....

"The government has made a hash of this across almost every index. I think that the fall out could be fairly large when people are told the this government's failures led to their loosing their grandparents and that in other countries that followed the science they had only a fraction of our deaths."

Stark.

the come back is we did follow the science (a week late). my data scientist chap told me a review panned Ferguson's code, inconsist results, variation upto 80k mortality each run. so who knows what happens with a different world.

 

[dazzer6666]

That’s my intention.

The USA is a difficult one to compare to European nations too. Effectively half a continent, with the megalopolis of NYC/NJ suffering like Paris, London and Madrid, whilst huge swathes in the centre are seemingly little scathed. Better imo to look at individual states.

Exactly......uk compares with NY state, not the US.

 

[Harry Wilson's tackle]

Numbers update, still steady fall in numbers and cases across the world (US and maybe Brazil excepted). Another couple of weeks before we see whether the easing is increasing cases (and 2 weeks later, deaths). So far, so good (but very slow).

 

[Weststander]

A useful graphic/statistic at the daily press briefing just then, where they synched (was this the first time?) actual deaths with covid versus suspected deaths with covid as at the exact same date, in this case 15th May.

46,371 versus 33,998, a difference of 36%, which can be used to gross-up as an estimate every daily total of deaths announced.

This is different from the higher still, total excess deaths, which includes those who’ve died up to today, those who’ve died from other causes but didn’t seek medical help possibly due to a fear of going to hospital just now and increased suicides.

 

[WATFORD zero]

From the latest ONS stats up to 15th May - Total fatalities 45,231.

These are ONS statistics where Hospital Fatalities have tested positive for Covid and outside of hospital (Home, Care Home, Other) where COVID-19 was mentioned on the death (ie may or may not be tested). A doctor can certify the involvement of COVID-19 based on symptoms and clinical findings – a positive test result is not required. Definitions of COVID-19 for deaths in Scotland and Northern Ireland are similar to England and Wales.

The total Excess deaths 59,228 up to 17 May. (No explanation of the difference of 15,000, but this was the difference asked about at PMQ two weeks ago, when it was 10,000).

Every measure now looks to be on a significant downward trend, except for care homes which is descending, albeit at a lower rate. As a result of the lower rate of reduction in care home the deaths over the last week reported have proportionally increased to 85% of the number hospital deaths (which is decreasing). I would hope this Care Home death rate would start to also trend downward more significantly over the next 2 weeks.

*edit*
Seeing [MENTION=21158]Weststander[/MENTION]'s post above, I missed the graphs yesterday.
Just found the slide - 33,998 positive Covid tested. 45,231 Covid on Death certificate, not tested. If that's true, it means 80K have died, so I guess they mean 11,233 have Covid on death certificate without a Covid test.

 

[Weststander]

From the latest ONS stats up to 15th May - Total fatalities 45,231.

These are ONS statistics where Hospital Fatalities have tested positive for Covid and outside of hospital (Home, Care Home, Other) where COVID-19 was mentioned on the death (ie may or may not be tested). A doctor can certify the involvement of COVID-19 based on symptoms and clinical findings – a positive test result is not required. Definitions of COVID-19 for deaths in Scotland and Northern Ireland are similar to England and Wales.

The total Excess deaths 59,228 up to 17 May. (No explanation of the difference of 15,000, but this was the difference asked about at PMQ two weeks ago, when it was 10,000).

Every measure now looks to be on a significant downward trend, except for care homes which is descending, albeit at a lower rate. As a result of the lower rate of reduction in care home the deaths over the last week reported have proportionally increased to 85% of the number hospital deaths (which is decreasing). I would hope this Care Home death rate would start to also trend downward more significantly over the next 2 weeks.

*edit*
Seeing [MENTION=21158]Weststander[/MENTION]'s post above, I missed the graphs yesterday.
Just found the slide - 33,998 positive Covid tested. 45,231 Covid on Death certificate, not tested. If that's true, it means 80K have died, so I guess they mean 11,233 have Covid on death certificate without a Covid test.

Digressing, almost under the radar, Spain went to a different (lower) method of reporting covid19 deaths on Sunday. Now just releasing confirmed as covid19 deaths, rather than those reported by hospitals, doctors and care homes. They took the opportunity to release a negative daily return.

Making comparisons between nations even more fraught.

I stand by the view that total excess deaths is the only comparable and real figure per nation.
[MENTION=396]WATFORD zero[/MENTION] always on the right lines for the UK. In time we should have reliable and up to date data for other nations.

 

[WATFORD zero]

Digressing, almost under the radar, Spain went to a different (lower) method of reporting covid19 deaths on Sunday. Now just releasing confirmed as covid19 deaths, rather than those reported by hospitals, doctors and care homes. They took the opportunity to release a negative daily return.

Making comparisons between nations even more fraught.

I stand by the view that total excess deaths is the only comparable and real figure per nation.
[MENTION=396]WATFORD zero[/MENTION] always on the right lines for the UK. In time we should have reliable and up to date data for other nations.

Hadn't seen that about Spain, bit late in the day to be changing basis isn't it ?

Not sure we'll ever get the truth from China, Russia, N Korea, but I think that bar a few exceptions we will know. That aside, I still find the graph halfway down this page useful for comparing countries, and it keeps getting updated, it's just the delays in reporting (I'm looking at you Italy and US !) that are frustrating.

https://www.ft.com/content/a26fbf7e-48f8-11ea-aeb3-955839e06441

 

[Weststander]

Hadn't seen that about Spain, bit late in the day to be changing basis isn't it ?

Not sure we'll ever get the truth from China, Russia, N Korea, but I think that bar a few exceptions we will know. That aside, I still find the graph halfway down this page useful for comparing countries, and it keeps getting updated, it's just the delays in reporting (I'm looking at you Italy and US !) that are frustrating.

https://www.ft.com/content/a26fbf7e-48f8-11ea-aeb3-955839e06441

The FT (on the free side of their pay-wall) has been a neat source of facts.

 

[WATFORD zero]

The FT (on the free side of their pay-wall) has been a neat source of facts.

Always liked the FT and it's because it comes from a consistent financial viewpoint, it's always interesting to see the financial markets perspective to various economic, health and political stories.

 

[Weststander]

Always liked the FT and it's because it comes from a consistent financial viewpoint, it's always interesting to see the financial markets perspective to various economic, health and political stories.

Same here, as well as The Economist and The Telegraph (it's second nature to me to ignore political editorials).

Annoyingly all three require not insignificant subscriptions these days, I'm holding out in not spending on this.

 

[nicko31]

Same here, as well as The Economist and The Telegraph (it's second nature to me to ignore political editorials).

Annoyingly all three require not insignificant subscriptions these days, I'm holding out in not spending on this.

Is there a place where you can pay a fee and get behind all these paywall at once?

 

[Chicken Run]

Is there a place where you can pay a fee and get behind all these paywall at once?

Support your local newsagent, she or he may even provide delivery service?

 

[dazzer6666]

Same here, as well as The Economist and The Telegraph (it's second nature to me to ignore political editorials).

Annoyingly all three require not insignificant subscriptions these days, I'm holding out in not spending on this.

I'm currently trying a year's online Telegraph subscription....hasn't quite worked out really though as primary reasons were a) having time to read when away/commuting and b) decent sports coverage. I'm obviously stuck at home with no sport to read about but still don't read it cover to cover every day even then. Doubt I'll renew as the cost is pretty high and not sure it's worth it over and above the free stuff available online.