Coronavirus outbreak

[Rocky]

View attachment 570243
Black and red lines have gone up for the whole of January, mind. Also note that the 'buff' line (average year) would go up in January by several hundred (flu etc in normal winter).

I have a problem with this representation of the impact of Covid. The FT reports that:

The jump in mortality rates suggests that since the pandemic began, the UK’s total excess deaths — the number above the previous five-year average — has risen to well over 100,000, according to a Financial Times model that brings official figures, which have a two-week lag, up to date.

https://www.ft.com/content/5f7b58fb-97ad-4fef-bbc9-b71d328c6700

Oke's figures seem to claim that the current number of deaths are roughly in line with the five year average. The FT disagrees with this. My money is on the FT here.

 

[Ajax Bay]

Oke's figures seem to claim that the current number of deaths are roughly in line with the five year average. The FT disagrees with this.

Don't think the FT figures are incompatible with the graph. Note what the 'y' axis plots. To get total deaths in 2020 from the graph I shared (from the Centre for Evidence Based Medicine, Oxford University) you have to integrate the two curves' difference over the year (365 days) - ie the net area under the black curve and above the 'buff' line. My money is on both of them for the joint 'win'.
Didn't one just love calculus bitd?

 

[Rocky]

Don't think the FT figures are incompatible with the graph. Note what the 'y' axis plots. To get total deaths in 2020 from the graph I shared (from the Centre for Evidence Based Medicine, Oxford University) you have to integrate the two curves' difference over the year (365 days) - ie the net area under the black curve and above the 'buff' line. My money is on both of them for the joint 'win'.
Didn't one just love calculus bitd?

I’d be wary about CEBM - its recent outputs haven’t been peer reviewed and its website is hosted on a private server. There is quite a lot of disquiet about some of their claims. Plus the data points seem to end in November...I can’t work out how they’ve continued the extrapolation.

 

[Ajax Bay]

the graph I shared from the Centre for Evidence Based Medicine, Oxford University

I’d be wary about CEBM - its recent outputs haven’t been peer reviewed and its website is hosted on a private server. There is quite a lot of disquiet about some of their claims.

https://www.phc.ox.ac.uk/research/oxford-centre-for-evidence-based-medicine
I tried to explore what one might be wary of but could not find a critical approach or expressions of disquiet. Could you offer a few pointers? In these fast-moving times, much stuff is not peer-reviewed by the time their conclusions are given air time. Apologies for my ignorance: what does having a website hosted on a "private server" suggest, to its detriment? Or is this more of an 'anti evidence-based medicine' issue, for you?
Data extrapolation - no can't help, sorry - this link which is updated daily, with provenance I guess suggests that this is 'real' and not a mere extrapolation form November:
https://www.cebm.net/covid-19/covid-19-death-data-in-england-daily-update/
Take it you agree that the CEBM data and the graph's presentation are entirely compatible with the FT numbers?

 

[Rocky]

https://www.phc.ox.ac.uk/research/oxford-centre-for-evidence-based-medicine
I tried to explore what one might be wary of but could not find a critical approach or expressions of disquiet. Could you offer a few pointers? In these fast-moving times, much stuff is not peer-reviewed by the time their conclusions are given air time. Apologies for my ignorance: what does having a website hosted on a "private server" suggest, to its detriment? Or is this more of an 'anti evidence-based medicine' issue, for you?
Data extrapolation - no can't help, sorry - this link which is updated daily, with provenance I guess suggests that this is 'real' and not a mere extrapolation form November:
https://www.cebm.net/covid-19/covid-19-death-data-in-england-daily-update/
Take it you agree that the CEBM data and the graph's presentation are entirely compatible with the FT numbers?

Have a look at how many peer reviewed journal articles its academics have published on Covid in 2020 (it’s pretty close to zero). Have a look at how many articles other academics from the Nuffield Department of General Practice have published - well into the 100s. Also have a look at the criticism of its director for advice given on mask wearing, testing and lockdown. There’s been much published on line and in the newspapers.

It may have a university link but it is hosted on a private server much to the chagrin of the university’s authorities.

 

[tom73]

Wonder if he's a friend of Carl's ?
As you've pointed out other parts have been going great work though out this and have some really high rate clinical knowledge and experience. Leading to some improved clinical outcomes and greater understanding of covid and shaping services for long covid.
It's a shame other parts have been so vocal about calling for other unproven ways out this mess. Leading to more mud in the waters that we can do without.

 

[Ajax Bay]

Wonder if he's a friend of Carl's ?

If you mean Prof Carl Heneghan, Dr Oke is a colleague of the CEBM director. No idea whether they are friends.
Not clear to me what toes are being stepped on here. I just shared a graph (which I will lay a bet is based on quality data) showing the distribution of deaths during 2020.
https://www.cebm.net/oxford-covid-19-evidence-service/

 

[Rocky]

If you mean Prof Carl Heneghan, Dr Oke is a colleague of the CEBM director. No idea whether they are friends.
Not clear to me what toes are being stepped on here. I just shared a graph (which I will lay a bet is based on quality data) showing the distribution of deaths during 2020.
https://www.cebm.net/oxford-covid-19-evidence-service/

The CEBM has consistently argued that the reported death stats are too high and that what we are experiencing is nothing different to 'normal death rates'. The original graph that was posted suggests that too with the 5 year average mostly tracking the 'all deaths' line. The red Covid line peaks in Mar-May but appears to be at zero throughout the summer. We know that large numbers of people were dying of Covid in December and January but there doesn't seem to be any data points on that graph. There have been nearly 100,000 excess deaths due to Covid (as per my FT reference). As I said in my first post, this appears to be misleading. Why that has been published like that on (a non-peer reviewed) website, we can only speculate.

Edit: what I struggle with is the reporting from the front line about ICUs overflowing and the need for temporary mortuaries and the CEBM claims that this is normal.

 

[Ajax Bay]

The CEBM has consistently argued that the reported death stats are too high and that what we are experiencing is nothing different to 'normal death rates'. The original graph that was posted suggests that too with the 5 year average mostly tracking the 'all deaths' line. The red Covid line peaks in Mar-May but appears to be at zero throughout the summer. We know that large numbers of people were dying of Covid in December and January but there doesn't seem to be any data points on that graph. There have been nearly 100,000 excess deaths due to Covid (as per my FT reference).

Thank you.
"The CEBM has consistently argued that the reported death stats are too high and that what we are experiencing is nothing different to 'normal death rates'." Does this graph suggest that? It doesn't to me. Look at the huge excess death spike in Mar-May and the excess deaths in December. That's your (FT's) 100,000, isn't it?
You queried:
1) "5 year average mostly tracking the 'all deaths' line". Why do you think that is not true for most of 2020's months?
2) "red Covid line peaks in Mar-May but appears to be at zero throughout the summer" Well, the death rate from COVID-19 (28 day criterion) averaged (my estimate from the gov.uk graph: https://coronavirus.data.gov.uk/details/deaths?areaType=nation&areaName=England ) about 20 a day from 1 Jul to 1 Oct, so on the CEBM graph that's close to zero, visually.
3) "large numbers of people were dying of Covid in December and January" Two observations: this graph is for 2020 (not this month) and the death rate (by death) in December only ticked above 400 on the 17th and much later. 400 in Nov/Dec is about what the CEBM graph shows. This graph may be the death rate by date reported in which case the up tick (above 400) is as late as 27 Dec.

(Better quality graph in my earlier post)

 

[Rocky]

Thank you.
"The CEBM has consistently argued that the reported death stats are too high and that what we are experiencing is nothing different to 'normal death rates'." Does this graph suggest that? It doesn't to me. Look at the huge excess death spike in Mar-May and the excess deaths in December. That's your (FT's) 100,000, isn't it?
You queried:
1) "5 year average mostly tracking the 'all deaths' line". Why do you think that is not true for most of 2020's months?
2) "red Covid line peaks in Mar-May but appears to be at zero throughout the summer" Well, the death rate from COVID-19 (28 day criterion) averaged (my estimate from the gov.uk graph: https://coronavirus.data.gov.uk/details/deaths?areaType=nation&areaName=England ) about 20 a day from 1 Jul to 1 Oct, so on the CEBM graph that's close to zero, visually.
3) "large numbers of people were dying of Covid in December and January" Two observations: this graph is for 2020 (not this month) and the death rate (by death) in December only ticked above 400 on the 17th and much later. 400 in Nov/Dec is about what the CEBM graph shows. This graph may be the death rate by date reported in which case the up tick (above 400) is as late as 27 Dec.

View attachment 570310
(Better quality graph in my earlier post)

I simply note that having the two lines tracking each other (apart from the spring uptick) suggests that nothing out of the ordinary is currently happening. Data from the ground suggests that is not the case and that CEBM for some reason is being disingenuous. Why that is happening, I don't know. For reasons I don't want to go into, I'm going to leave this discussion here.

 

[Ajax Bay]

CEBM - Deaths per week, by date of registration
Can see the effect (on registration) of the key public holidays every year (NY, Easter (varying and not in 2020), late May, late Aug, Christmas)
I think (but I don't know) that these data are from the ONS data: https://www.ons.gov.uk/peoplepopula...nalfiguresondeathsregisteredinenglandandwales
which also offers data (2020 COVID-19 deaths daily registered) to inform the previous graph I shared.

Irene M. Stratton MSc FFPH
Senior Statistician
Gloucestershire Retinal Research Group

 

[midlife]

Red line does seem to be an outlier.....

 

[Ajax Bay]

Red line does seem to be an outlier.....

Sadly it will be an outlier in 2021, at least for the first 4 months. Maybe will be below average after that, so an outlier then too.

 

[Milzy]

Sadly it will be an outlier in 2021, at least for the first 4 months. Maybe will be below average after that, so an outlier then too.

People think everything will be all tickety boo for the spring. Well it’s going to be death & lock downs on & off for most of the year in reality. The vaccines will only ease the situation not completely make it go away. It will take a full year to get most of the population done. 2022 should be more ‘normal’ apart from the economic damage.

 

[slowmotion]

People think everything will be all tickety boo for the spring. Well it’s going to be death & lock downs on & off for most of the year in reality. The vaccines will only ease the situation not completely make it go away. It will take a full year to get most of the population done. 2022 should be more ‘normal’ apart from the economic damage.

That was pretty much the conclusion that four college friends came to when we had a Zoom party earlier this evening. I wish I could, but I don't think a shot of a vaccine is going to wave some kind of magic wand over this mess.