COVID-19

Covid-19 discussion, bring your own statistics
Locked
sheldrake
After Pie
Posts: 1819
Joined: Fri Dec 20, 2019 2:48 am

Re: COVID-19

Post by sheldrake » Wed Oct 13, 2021 11:42 pm

I would like to blame Cummings for it.

OffTheRock
Fuzzable
Posts: 250
Joined: Mon Dec 02, 2019 6:52 pm

Re: COVID-19

Post by OffTheRock » Thu Oct 14, 2021 8:16 am

raven wrote:
Wed Oct 13, 2021 3:58 pm
Back to the +ve LFT, -ve PCR for a moment.
jdc wrote:
Sun Oct 10, 2021 4:52 pm
Couple of additional suggestions from Kit Yates - increased reporting, people faking a positive LFT result, or changes in test accuracy due to vaccination or demographic infected. (He also thinks faulty tests and new variant are possible explanations.)

More here, including a diagram based on 1% prevalence https://twitter.com/Kit_Yates_Maths/sta ... 6328343552
I like his diagram, much easier to see how the percentages shake out from a picture.

Kit Yates on Twitter PCR & LFT %.png

So, very roughly, for every 25 people with a +ve LFT, about 24 people really have Covid, and 1 doesn't. But 2 of those 25 people will have a subsequent -ve PCR, and we won't know which of them is infected.

I'm sure that he's right and there'll be a positive feedback loop of people reporting this on social media getting responses from other people it's happened to. But I don't know that it seems likely enough from those % to account for it happening to multiple people in the same family though.

If what I read yesterday online is accurate....

It's possible the LFT might be picking up another corona virus because the LFTs have been checked against 3 of the 4 common ones here and don't pick them up, but the 4th is too hard to grow in vitro so hasn't been checked & might be the issue.

Our PCR tests usually look at 3 points of RNA, so it seems unlikely a new variant would read as completely negative. You'd be more likely to see a drop out on one test-point I think, like there's been for variants before -- Delta iirc -- and that still counts as positive. Plus I think we're still doing a fair bit of sequencing from PCRs so I'd hope a new variant would get picked up that way.
Bêtes theorem and social media does explain what is happening in most of the country. Especially as the more testing is going on in secondaries and the higher the prevalance is the more you’d expect to see. Some areas of the SW are completely atypical though. Here’s a thread with some data from the 9 ‘suspicious areas’ in the SW. https://mobile.twitter.com/ArtySmokesPS ... 0837458949

Based on this and some anecdotal evidence of schools in Kettering/Corby seeing very few cases of this and schools in NE Somerset seeing dozens of cases but very little covid and people in the area testing at a testing centre and getting a postal testing kit (which does come back positive) I’d say it’s looking like a lab issue. I’m not sure how long it will take UKHSA to figure this out though.

User avatar
shpalman
Light of Blast
Posts: 5831
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond

Re: COVID-19

Post by shpalman » Thu Oct 14, 2021 11:30 am

sheldrake wrote:
Wed Oct 13, 2021 10:33 pm
shpalman wrote:
Wed Oct 13, 2021 10:10 pm
So for example the worst week in the first wave was that of the 17th of April in which there were 21144 deaths when normally about 10000 would have been expected; only about 8000 were even attributed to covid on the death certificate, but there were only about 5000 deaths per week in the covid stats just because someone won't count as a covid death unless they tested positive for covid, and the testing capacity just wasn't there yet. So yeah some of those tens of thousands of deaths might have been avoided by locking down earlier.
1) 'Excess deaths' depend on what you take as a 'normal' level of deaths. Previous 5 years we had unusually low age-adjusted mortality compared to the norms going back to 1990. You can't assume that all, or even most, deaths above the previous 5-year average were really the result of untested Covid.


Do you have week-by-week data going back to 1990 showing how it was 'normal' for 20,000 people to die per week at the beginning of April only as compared to more like 10,000 per week the rest of the year?
sheldrake wrote:
Wed Oct 13, 2021 10:33 pm
2) Deaths that occur even when somebody tests positive to Covid aren't necessarily primarily caused by Covid. I think the cause of admission to hospital (e.g. admitted for Covid, or unknown respiratory) would be the best thing to look at, don't know if it's possible to dig that out.
Well, there's whether or not covid was mentioned on the death certificate. But you can't really go by hospital admissions either, since in the first wave older people quickly went from "seem to be doing fine" to "too late to do anything" and there wasn't really considered much point in taking them to hospital. Doctors didn't really know what to do except prone intubation and older patients were considered not strong enough to survive it. Survival rates in intensive care were about 50% at that stage. They've since developed better protocols and with fewer patients being admitted have the chance to actually apply them.
molto tricky

User avatar
shpalman
Light of Blast
Posts: 5831
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond

Re: COVID-19

Post by shpalman » Thu Oct 14, 2021 11:42 am

But some of those excess deaths will have been from treatable things which weren't covid, but there wasn't the chance to treat them because the hospitals were full of the covids. Because the first lockdown was ordered two weeks too late.
molto tricky

sheldrake
After Pie
Posts: 1819
Joined: Fri Dec 20, 2019 2:48 am

Re: COVID-19

Post by sheldrake » Thu Oct 14, 2021 11:55 am

shpalman wrote:
Thu Oct 14, 2021 11:30 am


Do you have week-by-week data going back to 1990 showing how it was 'normal' for 20,000 people to die per week at the beginning of April only as compared to more like 10,000 per week the rest of the year?
No, I've only seen annual age-adjusted death data. 2020 was below the norms for 2008 and every year going back to 1990 though.

raven
Snowbonk
Posts: 458
Joined: Mon Nov 11, 2019 8:58 pm

Re: COVID-19

Post by raven » Thu Oct 14, 2021 4:08 pm

@Woodchopper & OffTheRock -

Yes, it does look like it's a testing issue. I don't know if all the PCR labs use the three-point test - iirc at one point some were using a 2point one, so I wondered if it might be that. Definitely looks like something pretty dramatic is happening, whether it's at the lab stage or the swabbing stage. Do all the SW swabs go to one lab though? I thought the mega labs at Leamington & Milton Keynes took samples from all over.

raven
Snowbonk
Posts: 458
Joined: Mon Nov 11, 2019 8:58 pm

Re: COVID-19

Post by raven » Thu Oct 14, 2021 4:25 pm

In response to some of sheldrake's points:

1) Behavourial science indicated people would be less compliant than they were, so perhaps TPTB thought a firm tone was needed, but certainly in early 2020 I remember government messaging as reassuring rather than alarmist. It's hard for me to judge how much supression there was because I actively sought information, but to be fair to government simple and direct public health messages are more effective than lengthy discussion of uncertainties. (Acknowledging that uncertainty, as Sturgeon tends to do, is better than ignoring it though.)

2) Risk is tricky to understand, and no-one wants to die, so it's always going to be politically hard to 'allow' lots of people to die, particularly when so many Covid deaths were horrible and lonely. I think we might be at the point where we're accepting of a certain level of Covid death now though.

3) I guess the Nightingales were unused because there weren't enough people to staff them & because the NHS did some massive rejigging and increased capacity elsewhere by 53%. You also have to factor in that during the first wave a significant number of Covid patients stayed at home (as they were told to) and died there so didn't take up beds.

4) Ok, I'll bite. Yes, waves of cases peak and drop away before everybody has caught it. That happened in 1918 with flu too. Cliquey social networks and/or changing behaviour is probably the explanation for that. The reason you'd want lockdown to be a government mandated thing is because some employers are gits and will demand workers come in to work even when it's possible to do their jobs from home. Similarly, closing state schools.

5) Wrt to discharging care home residents tout de suite. Hospitals tended to be sources of transmission in a pandemic, often a significant one. So what do you, back in early 2020, with elderly asymptomatic inpatients who are very much at risk if they stay on a ward? Do you keep them at risk while you wait for a test, if you can even get one done quick enough? Or do you put the rest of their care home at risk by discharging them? That's not an easy decision.

It's a similar balancing act with visiting patients at home & cutting down on face-to-face GP appointments - it's not just the risk to that GP getting sick, or the resulting loss of staff hours to the practise, it's also the risk of transmission to subsequent patients in that infectious yet-not-symptomatic interval.

6) On death rate back to 1990, you might want to look at what the ONS say about their own data. See here, and here and here.
tl, dr version: deaths fell from 1990 to 2010 because of ongoing improvements in lifestyle/health (mostly men's life expectancy catching up with womens I think). That trend slowed after 2010, and in fact as a straight number deaths rose from 2015, perhaps because of 'flu, or austerity. So comparing 2020 to the last five years makes some sense.
Screenshot 202Causes of death over 100 years - Office for National Statistics.png
Screenshot 202Causes of death over 100 years - Office for National Statistics.png (22.45 KiB) Viewed 383 times

sheldrake
After Pie
Posts: 1819
Joined: Fri Dec 20, 2019 2:48 am

Re: COVID-19

Post by sheldrake » Thu Oct 14, 2021 5:04 pm

It's a similar balancing act with visiting patients at home & cutting down on face-to-face GP appointments - it's not just the risk to that GP getting sick, or the resulting loss of staff hours to the practise, it's also the risk of transmission to subsequent patients in that infectious yet-not-symptomatic interval.
I think this has to be balanced against the fact that people asking to see a doctor are likely to be sick with something else requiring treatment. Covid is far from the only thing that can kill you, and it's not even the most likely condition to kill most people. This is one of the dangers of the single-frame focus I talked about.
6) On death rate back to 1990, you might want to look at what the ONS say about their own data. See here, and here and here.
tl, dr version: deaths fell from 1990 to 2010 because of ongoing improvements in lifestyle/health (mostly men's life expectancy catching up with womens I think). That trend slowed after 2010, and in fact as a straight number deaths rose from 2015, perhaps because of 'flu, or austerity. So comparing 2020 to the last five years makes some sense.
Their analysis doesn't really give a cause for the changes in mortality over time, it just notes they're happening. The age-adjusted mortality was higher as recently as 2008 - is there really strong evidence that this is all due to lifestyle changes and the last 5 years represent a 'new normal', or could it be that the last 5 years just had coincidentally low mortality due to different flu strains, or an increase in flu vaccine uptake etc.. etc.. ?

User avatar
shpalman
Light of Blast
Posts: 5831
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond

Re: COVID-19

Post by shpalman » Thu Oct 14, 2021 6:46 pm

I assume you're looking at the Annual deaths and mortality rates, 1838 to 2020 (provisional) spreadsheet.

The number of deaths is corrected both for the increasing population, and then for the increasing age of the population. The figures for 2020 are still only provisional.
prov-age-std-death-rate.png
prov-age-std-death-rate.png (21.66 KiB) Viewed 357 times
We can see it steadily falling until about 2010, and then it stays fairly stable, but the uptick for 2020 is visible even on this scale.

In terms of the absolute numbers of deaths it looks like this since 1990:
number-of-deaths-by-year.png
number-of-deaths-by-year.png (9.29 KiB) Viewed 357 times
So I don't buy that 2020 is regression to the mean. It's something like 50,000 more deaths than you would have expected by continuing the trend and allowing for the kinds of year-to-year variations you see recently (allowing for 2019 having been a bit lower than we might have expected).

The official cumulative total of "Deaths within 28 days of positive test by date of death" was 76,448 by the 31st of December 2020. "Daily deaths with COVID-19 on the death certificate by date of death" reached 94,998 by that date.

But we can look at weekly deaths back to 2010 and compare it to 2020, and what we have of 2021 so far. Maybe overlay the official covid deaths on it.

It's been a while since we did that.

Image
molto tricky

sheldrake
After Pie
Posts: 1819
Joined: Fri Dec 20, 2019 2:48 am

Re: COVID-19

Post by sheldrake » Thu Oct 14, 2021 7:00 pm

shpalman wrote:
Thu Oct 14, 2021 6:46 pm
I assume you're looking at the Annual deaths and mortality rates, 1838 to 2020 (provisional) spreadsheet.
No, I'm looking at the link I posted earlier which shows age adjusted deaths per capita, in which 2020 is just a return to 2009 levels and still lower than all previous years to 1990
https://www.ons.gov.uk/aboutus/transpar ... 51wKzm7uiQ

Here's a plot of age-adjusted mortality per 100,000 over 30 years
Screen Shot 2021-10-14 at 8.09.33 PM.png
Screen Shot 2021-10-14 at 8.09.33 PM.png (90.91 KiB) Viewed 339 times
I don't think we can use the 'follow the trend' method, need to look at good breakdown of causes of death to be sure it's not what I suggested before.
Last edited by sheldrake on Thu Oct 14, 2021 7:11 pm, edited 1 time in total.

User avatar
mediocrity511
Snowbonk
Posts: 408
Joined: Mon Nov 11, 2019 2:16 pm

Re: COVID-19

Post by mediocrity511 » Thu Oct 14, 2021 7:09 pm

raven wrote:
Thu Oct 14, 2021 4:08 pm
@Woodchopper & OffTheRock -

Yes, it does look like it's a testing issue. I don't know if all the PCR labs use the three-point test - iirc at one point some were using a 2point one, so I wondered if it might be that. Definitely looks like something pretty dramatic is happening, whether it's at the lab stage or the swabbing stage. Do all the SW swabs go to one lab though? I thought the mega labs at Leamington & Milton Keynes took samples from all over.
There's some discussion of people in the SW testing negative from drive through PCRs but positive via postal. Might the mega labs take all the postal stuff for example? Obviously testing again later during the course of the illness might mean higher viral load at that point.

User avatar
shpalman
Light of Blast
Posts: 5831
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond

Re: COVID-19

Post by shpalman » Thu Oct 14, 2021 7:12 pm

sheldrake wrote:
Thu Oct 14, 2021 7:00 pm
shpalman wrote:
Thu Oct 14, 2021 6:46 pm
I assume you're looking at the Annual deaths and mortality rates, 1838 to 2020 (provisional) spreadsheet.
No, I'm looking at the link I posted earlier which shows age adjusted deaths per capita, in which 2020 is just a return to 2009 levels and still lower than all previous years to 1990
https://www.ons.gov.uk/aboutus/transpar ... 51wKzm7uiQ
It's the same numbers
sheldrake wrote:
Thu Oct 14, 2021 7:00 pm
I don't think we can use the 'follow the trend' method, need to look at good breakdown of causes of death to be sure it's not what I suggested before.
The spreadsheets at the weekly deaths link I posted do that.
molto tricky

sheldrake
After Pie
Posts: 1819
Joined: Fri Dec 20, 2019 2:48 am

Re: COVID-19

Post by sheldrake » Thu Oct 14, 2021 7:14 pm

shpalman wrote:
Thu Oct 14, 2021 7:12 pm

It's the same numbers
Look at the plot I've posted where it's not just deaths, but age adjusted per-capita. Less dramatic looking, wouldn't you say? Remember life in 2008, 2005? You probably don't remember it being a frightening time requiring urgent govt. intervention in people's daily lives or unusual measures to try and strong-arm them into taking medications. More death then though, just less media hysteria.
sheldrake wrote:
Thu Oct 14, 2021 7:00 pm
I don't think we can use the 'follow the trend' method, need to look at good breakdown of causes of death to be sure it's not what I suggested before.
The spreadsheets at the weekly deaths link I posted do that.
Sure. I still don't think that's a good way to analyse this, for the reasons I gave a few posts ago.

User avatar
shpalman
Light of Blast
Posts: 5831
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond

Re: COVID-19

Post by shpalman » Thu Oct 14, 2021 8:54 pm

sheldrake wrote:
Thu Oct 14, 2021 7:14 pm
shpalman wrote:
Thu Oct 14, 2021 7:12 pm

It's the same numbers
Look at the plot I've posted where it's not just deaths, but age adjusted per-capita. Less dramatic looking, wouldn't you say? Remember life in 2008, 2005? You probably don't remember it being a frightening time requiring urgent govt. intervention in people's daily lives or unusual measures to try and strong-arm them into taking medications. More death then though, just less media hysteria.
This

Image

is the same data as this

Image

just different axes. Same uptick at the end for 2020 which is easier to see if your version because you only go back to 1990 rather than 1942.

(I also plotted, in the second graph, the actual death data so we could compare those numbers to the covid deaths data.)

Bear in mind that uptick in deaths was despite locking down hard (if late). None of us can really predict how bad things could have been without a government-mandated lockdown (people would have reacted anyway) but in the second wave the CFR was about 2% and even if you assume we were only picking up about 20% of cases (the rate at which tests were coming back positive peaked at about 17% in November 2020) that would still be an IFR of about 0.4% and the possibility of a couple of hundred thousand people dying earlier than they otherwise would have. (As compared to the normal number of deaths in a year the range of 500,000-600,000.)

Or maybe your assertion that the death rate has been a bit low over the past few years is implying that some people have already lived longer than they should have.
sheldrake wrote:
Thu Oct 14, 2021 7:14 pm
sheldrake wrote:
Thu Oct 14, 2021 7:00 pm
I don't think we can use the 'follow the trend' method, need to look at good breakdown of causes of death to be sure it's not what I suggested before.
The spreadsheets at the weekly deaths link I posted do that.
Sure. I still don't think that's a good way to analyse this, for the reasons I gave a few posts ago.
What's not a good way to analyse this? Tables 12a-c and 13a-c in this excel spreadsheet at the weekly deaths link I posted, which break down causes of death?
Last edited by shpalman on Thu Oct 14, 2021 8:56 pm, edited 1 time in total.
molto tricky

Herainestold
Dorkwood
Posts: 1422
Joined: Mon Nov 25, 2019 1:23 pm

Re: COVID-19

Post by Herainestold » Thu Oct 14, 2021 8:55 pm

Trevor Bedford thread on Covid becoming endemic

https://twitter.com/trvrb/status/1448297981279617027
Trevor Bedford
@trvrb
·
13 Oct
I've meaning to write a "COVID endgame" thread for a while and I apologize this is somewhat delayed compared to media interviews like https://science.org/content/article/wil ... mark-clues
and https://statnews.com/2021/09/20/winter- ... son-looms/ and to recent seminars like https://youtube.com/watch?v=VErVD_H1BZ0
. 1/17
Get vaccinated. Wear your mask. Keep distancing. Test twice a week. Work from home. Don't mix households. Avoid crowds. Open your windows. Don't go dancing. Get your booster.

sheldrake
After Pie
Posts: 1819
Joined: Fri Dec 20, 2019 2:48 am

Re: COVID-19

Post by sheldrake » Thu Oct 14, 2021 9:05 pm

shpalman wrote:
Thu Oct 14, 2021 8:54 pm

Look at the plot I've posted where it's not just deaths, but age adjusted per-capita. Less dramatic looking, wouldn't you say? Remember life in 2008, 2005? You probably don't remember it being a frightening time requiring urgent govt. intervention in people's daily lives or unusual measures to try and strong-arm them into taking medications. More death then though, just less media hysteria.
Bear in mind that uptick in deaths was despite locking down hard (if late).
Yes, but it's a tiny uptick that still looks like 'normal for 2009'. I don't remember 2009 being a particularly scary year, do you ?
None of us can really predict how bad things could have been without a government-mandated lockdown (people would have reacted anyway)
I beg to differ. We do have countries and US states to look at where there was no govt. mandated lockdown (and by lockdown, I'm talking about the kind of intense measures we saw in the UK, or Australia, not what they did in Sweden, which is routinely cited as an example of 'not locking down' despite having a rule of 8 in restaurants etc.. for a brief period).
Or maybe your assertion that the death rate has been a bit low over the past few years is implying that some people have already lived longer than they should have
I wouldn't say anything that dismissive of human life. I would simply say that what we've seen in 2020 was no more severe than what we considered 'normal' in 2009, or 'less severe than normal' for most preceeding years. It may be that there were multiple flu strains or other health problems at work in the population before 2009 that we were comfortable largely ignoring whereas this disease, which seems a less dangerous than driving, or junk food, has provoked a panicked reaction that's being used as a wedge to introduce mandatory digital IDs and travel restrictions etc..
What's not a good way to analyse this?
'Follow the trend' doesn't sit well with me yet because in 30 years of data, it could be 2010-2019 which are an abnormal quiet spot as far as various pathogens or other common causes of death go.

User avatar
shpalman
Light of Blast
Posts: 5831
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond

Re: COVID-19

Post by shpalman » Thu Oct 14, 2021 10:46 pm

Ah so here's what changed in 2010 when the downward trend stopped https://www.theguardian.com/society/202 ... five-years
molto tricky

sheldrake
After Pie
Posts: 1819
Joined: Fri Dec 20, 2019 2:48 am

Re: COVID-19

Post by sheldrake » Thu Oct 14, 2021 11:01 pm

Please stop treating Guardian political editorial as scientifically useful. This is obviously nonsense. The social sciences are full of ideologically obsessive morons who treat simple chronological sequencing of a few variables as evidence of causality. Age-adjusted mortality was even higher before 2010. Austerity doesn't explain that.

User avatar
Gfamily
Stummy Beige
Posts: 3148
Joined: Mon Nov 11, 2019 1:00 pm
Location: NW England

Re: COVID-19

Post by Gfamily » Thu Oct 14, 2021 11:21 pm

sheldrake wrote:
Thu Oct 14, 2021 11:01 pm
Please stop treating Guardian political editorial as scientifically useful. This is obviously nonsense. The social sciences are full of ideologically obsessive morons who treat simple chronological sequencing of a few variables as evidence of causality. Age-adjusted mortality was even higher before 2010. Austerity doesn't explain that.
If age adjusted mortality was falling before 2010 and rising after 2010, given that age only increases at 1 year per year, then something needs explaining.

Have you ever considered asking yourself "might I be wrong?" before posting?
My avatar was a scientific result that was later found to be 'mistaken' - I rarely claim to be 100% correct
ETA 5/8/20: I've been advised that the result was correct, it was the initial interpretation that needed to be withdrawn
Meta? I'd say so!

sheldrake
After Pie
Posts: 1819
Joined: Fri Dec 20, 2019 2:48 am

Re: COVID-19

Post by sheldrake » Thu Oct 14, 2021 11:31 pm

Gfamily wrote:
Thu Oct 14, 2021 11:21 pm

If age adjusted mortality was falling before 2010 and rising after 2010, given that age only increases at 1 year per year, then something needs explaining.

Have you ever considered asking yourself "might I be wrong?" before posting?
Absolutely, but that article was an extremely shallow, facile analysis of the kind that people who've started with a conclusion based on their own biases and then work backwards to try and make reality fit them would do. It's transparent politicking. There are a large number of possible explanations for mortality falling before 2010 and then flattening out, which the study did not evaluate. It's pretty much on a par with the old 'Global temps are rising because the number of pirates has gone down' joke. You all know we can do better than that.

User avatar
Gfamily
Stummy Beige
Posts: 3148
Joined: Mon Nov 11, 2019 1:00 pm
Location: NW England

Re: COVID-19

Post by Gfamily » Fri Oct 15, 2021 12:10 am

sheldrake wrote:
Thu Oct 14, 2021 11:31 pm
Gfamily wrote:
Thu Oct 14, 2021 11:21 pm

If age adjusted mortality was falling before 2010 and rising after 2010, given that age only increases at 1 year per year, then something needs explaining.

Have you ever considered asking yourself "might I be wrong?" before posting?
Absolutely, but that article was an extremely shallow, facile analysis of the kind that people who've started with a conclusion based on their own biases and then work backwards to try and make reality fit them would do. It's transparent politicking. There are a large number of possible explanations for mortality falling before 2010 and then flattening out, which the study did not evaluate. It's pretty much on a par with the old 'Global temps are rising because the number of pirates has gone down' joke. You all know we can do better than that.
"A mind so open that reason has fallen out"
My avatar was a scientific result that was later found to be 'mistaken' - I rarely claim to be 100% correct
ETA 5/8/20: I've been advised that the result was correct, it was the initial interpretation that needed to be withdrawn
Meta? I'd say so!

User avatar
shpalman
Light of Blast
Posts: 5831
Joined: Mon Nov 11, 2019 12:53 pm
Location: One step beyond

Re: COVID-19

Post by shpalman » Fri Oct 15, 2021 6:40 am

sheldrake wrote:
Thu Oct 14, 2021 11:01 pm
Please stop treating Guardian political editorial as scientifically useful. This is obviously nonsense. The social sciences are full of ideologically obsessive morons who treat simple chronological sequencing of a few variables as evidence of causality. Age-adjusted mortality was even higher before 2010. Austerity doesn't explain that.
Age-adjusted mortality was falling, and then it stopped falling.

However, I'll share your annoyance with the Guardian reporting on this without linking to the publications, which may not even be online at BMJ Open yet (that's from a year ago so that's not it).

They also mention something from Lancet Public Health which might be that for all I know, at least the authors' affiliations seem to be ICL.
In 2002–06 and 2006–10, all but a few (0–1%) MSOAs* had a life expectancy increase for female and male sexes. In 2010–14, female life expectancy decreased in 351 (5·2%) of 6791 MSOAs. By 2014–19, the number of MSOAs with declining life expectancy was 1270 (18·7%) for women and 784 (11·5%) for men. The life expectancy increase from 2002 to 2019 was smaller in MSOAs where life expectancy had been lower in 2002 (mostly northern urban MSOAs), and larger in MSOAs where life expectancy had been higher in 2002 (mostly MSOAs in and around London). As a result of these trends, the gap between the first and 99th percentiles of MSOA life expectancy for women increased from 10·7 years (95% credible interval 10·4–10·9) in 2002 to reach 14·2 years (13·9–14·5) in 2019, and for men increased from 11·5 years (11·3–11·7) in 2002 to 13·6 years (13·4–13·9) in 2019.

Interpretation
In the decade before the COVID-19 pandemic, life expectancy declined in increasing numbers of communities in England. To ensure that this trend does not continue or worsen, there is a need for pro-equity economic and social policies, and greater investment in public health and health care throughout the entire country.
* - "middle-layer super output areas"
molto tricky

bagpuss
Dorkwood
Posts: 1088
Joined: Tue Nov 12, 2019 12:10 pm

Re: COVID-19

Post by bagpuss » Fri Oct 15, 2021 6:43 am

Back to the +ve LFT / - ve PCR again, looks like it's definitely a lab issue. This article only mentions one testing site but the woman from Berks talking on the Today programme just now was basically saying it was a lab issue while carefully only talking about that one testing site because that's her area of responsibility while labs aren't. Sounds like there'll be confirmation of the lab issue later today.

User avatar
bob sterman
Catbabel
Posts: 714
Joined: Mon Nov 11, 2019 10:25 pm
Location: Location Location

Re: COVID-19

Post by bob sterman » Fri Oct 15, 2021 7:42 am

sheldrake wrote:
Thu Oct 14, 2021 9:05 pm
shpalman wrote:
Thu Oct 14, 2021 8:54 pm
Bear in mind that uptick in deaths was despite locking down hard (if late).
Yes, but it's a tiny uptick that still looks like 'normal for 2009'. I don't remember 2009 being a particularly scary year, do you ?
"Tiny uptick"? Sounds like something Boris would say while being indiscreet.

Here are the EUROMOMO excess mortality z-scores for England in recent years.
uptick.jpg
uptick.jpg (61.24 KiB) Viewed 244 times
I'm not sure I've ever heard of a z-score > 30 being described as a "tiny" effect.

Maybe you should write to EUROMOMO and tell them their methodology is wrong and they should be calculating everything with reference to 2009?

https://www.euromomo.eu/how-it-works/rationale/

User avatar
Woodchopper
Light of Blast
Posts: 4356
Joined: Sat Oct 12, 2019 9:05 am

Re: COVID-19

Post by Woodchopper » Fri Oct 15, 2021 8:21 am

NHS Test and Trace has suspended testing operations provided by Immensa Health Clinic Ltd at its laboratory in Wolverhampton.

It follows an investigation into reports of people receiving negative PCR test results after testing positive with a lateral flow test.

Investigations are under way into the precise cause but NHS Test and Trace estimates that around 400,000 samples have been processed through the lab.

The vast majority of these will have been negative results, but an estimated 43,000 people may have been given incorrect negative PCR test results between 8 September and 12 October, mostly in south-west England.
https://www.bbc.com/news/live/uk-58922518

So the LFT PCR issue could be due to a dodgy lab. That's bad but preferable to a new variant.

Locked