Coronavirus - General Chat - No statistics

[tjh290633]

The Wetherspoons I have my breakfast in has the NHS QR Code system and the pieces of paper. It is not overcrowded in the morning.

The same with ours for Curry Club lunch on Thursdays. No crowding, QR code on the table, and contact sheets if required.

TJH

If required by whom?

How would one know if NHS Track and Trace will be requiring them in a week?

Required if you don't use the QR code.

TJH

[Lootman]

The same with ours for Curry Club lunch on Thursdays. No crowding, QR code on the table, and contact sheets if required.

If required by whom?

How would one know if NHS Track and Trace will be requiring them in a week?

Required if you don't use the QR code.

?
Yes but the question was: Required by whom?

The pub? The government? The local council?

And what if anything happens if I decide I do not wish to give that information in any form?

[scotia]

As people are increasingly coming to see that CV is nothing like as dangerous as it's been portrayed it's inevitable that their willingness to undertake any exercise that might put them in house arrest for 14 days will rapidly diminish.

If you look at the most recent official report on the Covid-19 pandemic in England
https://www.imperial.ac.uk/media/imperial-college/institute-of-global-health-innovation/Ward-et-al-120820.pdf
I quote
Infection Fatality Ratio
The estimated community IFR (excluding care homes) was 0.9% (0.86, 0.94), increasing with age and
higher in males (1.07%, 1.00, 1.15) than females (0.71%, 0.67, 0.75) (Table 3). Sensitivity analyses
indicate an IFR as high as 1.58% (1.51%, 1.65%) if excess rather than COVID-specific deaths are used
and care home deaths are included (Table S2).
So getting a Covid-19 infection seems like a dangerous pursuit to me - but then I don't plan to climb Mount Everest which may put my risk appetite in perspective.

and the death rate has now actually been lower than normal for the past couple of months.

Of course it is. There was a large spike in the deaths among aged and infirm during the pandemic, and a fraction of these could have been expected to die within a few months. So we should see a slightly lower death rate during the following months once the initial spike had been quenched by lockdown. So that statistic cannot be interpreted that the virus is somehow less dangerous. However the encouraging news from the hospitals is that they have learned substantially from experience, and their treatments of Covid-19 now seem to be more effective than in the past. With the current small numbers of infections present in the different age cohorts it is difficult to quote numbers - but it looks like the mortality rate from a Covid-19 infection is certainly reducing. Good news - especially if the public also take sensible measures to prevent the virus multiplying.

Many people are therefore taking the view that the rules are increasingly pointless and based on knee-jerk reaction rather than scientific evidence, and substituting their own judgment as to what's safe and what's not.

maybe you should be out there shouting the message to the public - read the scientific evidence.
and I can't resist a little gentle dig - tell them its definitely not 5G that's to blame

[johnhemming]

I quote
Infection Fatality Ratio
The estimated community IFR (excluding care homes) was 0.9% (0.86, 0.94), increasing with age and
higher in males (1.07%, 1.00, 1.15) than females (0.71%, 0.67, 0.75) (Table 3). Sensitivity analyses
indicate an IFR as high as 1.58% (1.51%, 1.65%) if excess rather than COVID-specific deaths are used
and care home deaths are included (Table S2).

Also in the report

In estimating the IFR, we may have underestimated the number
of infected individuals (leading to higher estimates of IFR), as a result of weakened or absent
antibody response in some people, and waning antibody over time.

[redsturgeon]

https://www.axios.com/trump-covid-olean ... 13d41.html

Oh dear, Trump seems to be pushing the latest miracle cure for Covid!

John

[Mike4]

I'm perplexed about how it is possible for any agency or body to be quoting an IFR, given we still don't know what proportion of people who are infected, are infected asymptomatically.

Or maybe we do. Can anyone enlighten me please with some good quality peer reviewed research?

Much obliged.

[scotia]

I'm perplexed about how it is possible for any agency or body to be quoting an IFR, given we still don't know what proportion of people who are infected, are infected asymptomatically.

Or maybe we do. Can anyone enlighten me please with some good quality peer reviewed research?

Much obliged.

Publishing a peer reviewed paper is a long process. First a paper is written -probably with a team of authors who have contributed, and who want to make sure that it reflects their views. This takes time. Then it is sent off to a reputable Journal, and its Editorial team read it to see if it may be worthy of their attention, and if so they send it to a number of possibly knowledgeable referees - and they make a considerable time to report back and they may also request alterations and clarifications, which the authors then need to make. And in this way its possible for a year to pass before publication takes place (at least this is how it used to be in my day!).
So with something as urgent as Covid-19, corners need to be cut - and consequently we may have to make do (at least temporarily) with something less well reviewed.
The most recent comprehensive study of the Pandemic in England is the report I took my quotes from concerning IFR. As you probably are aware they based their infection numbers on the number of persons identified with antibodies and they provided evidence, with a high degree of confidence that such antibodies had not faded sufficiently over several months to make them undetectable. There are a number of caveats - its worth looking at them.

But how do you identify asymptomatic infections? Johnhemming provided a reference which summarised work from a Chinese study.
https://www.cidrap.umn.edu/news-perspective/2020/06/chinese-study-antibodies-covid-19-patients-fade-quickly
It appears that the Chinese group had no problems investigating asymptomatic patients - since they had roughly similar antibody levels to symptomatic patients, and the levels faded with time at roughly similar rates in both cases.
Its difficult to know what (if any) reliance to place on this source - but if (and I stress if) it is correct then it suggests that antibody levels are a feature in all Covid-19 infections, and the estimates of IFR in England are reasonably robust
Edit - I have added roughly (twice) in my comments on the Chinese paper - and I'm trying to find a first-hand source
First Hand Source https://www.nature.com/articles/s41591-020-0965-6

[scotia]

https://www.nature.com/articles/s41591-020-0965-6
I have now had a read-through of the above reference and its definitely worth reading. I wish I had seen this when it was published two months ago.
From a sample of 178 persons identified as being infected, 37 proved to be asymptomatic - and were studied. For comparison 37 patients with mild symptomatic symptoms were independently selected from a similar age/sex cohort. The study focussed largely on the antibodies detected in both sets of patients.
The median duration of viral shedding in asymptomatic patients was longer than in the symptomatic patients (but with large spreads)
The median antibody levels in the asymptomatic patients were smaller than in the symptomatic patients - but again the spreads were large and overlapping
The decay rate of antibody levels in both sets was similar.
The authors stated that the ratio of 37 asymptomatic infections in a sample of 178 infections may not necessarily be an accurate estimate of the ratio in the entire population "-owing to the fact that asymptomatic infections were identified from those who were at high risk for infection"
The group has a further publication on identifying Covid-19 by antibodies
https://www.nature.com/articles/s41591-020-0897-1
it is also worth a read and reports that antibody tests can succeed in detecting Covid-19 when other methods fail.

My opinion after reading these papers is that an antibody test is a reliable way of identifying Covid-19 - both asymptomatic and symptomatic. So in the UK, from our sampling of antibodies, we have a good estimate of the total number of Covid-19 infections - under 10% with regional variations. And the IFR of around 1% is in the correct ballpark.

[johnhemming]

I'm perplexed about how it is possible for any agency or body to be quoting an IFR, given we still don't know what proportion of people who are infected, are infected asymptomatically.

The Imperial College paper linked to above uses antibodies to work back to an IFR given community deaths. They recognise, however, that there are uncertainties with this. My view is that their figure is worse than useless because of the uncertainties being all in one direction.

Another problem with IFR is that it appears that how people are cared for affects the outcome. Hence if people get a larger viral load via hospital treatment, for example, then the IFR will go up. Similarly if people in care homes are systematically infected as happened in New York then that causes more deaths. North Italy had a lot of additional deaths through this vector.

I did see a report which indicated 20% of infections were caught in hospital in England. However, on a quick search I have only seen the figure of 12.5% reported.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7372282/

[johnhemming]

My opinion after reading these papers is that an antibody test is a reliable way of identifying Covid-19 - both asymptomatic and symptomatic. So in the UK, from our sampling of antibodies, we have a good estimate of the total number of Covid-19 infections - under 10% with regional variations. And the IFR of around 1% is in the correct ballpark.

It is known that a proportion of people's antibodies fade and that some people don't generate antibodies. Hence any calculation that uses the presence of antibodies as the only test of infection will inevitably have too small a figure. Hence that could only produce an upper bound of the actual figure for Infections Fatality Rate.

I accept that we won't agree about this. I think we both have some form of scientific and mathematical training. In a sense that demonstrates that people with scientific training can some out with different answers. Similarly Imperial College and Oxford University tend to come out with different answers. Imperial was, in fact, my second choice for university back in the 1970s, but I got my first choice.

[scotia]

My opinion after reading these papers is that an antibody test is a reliable way of identifying Covid-19 - both asymptomatic and symptomatic. So in the UK, from our sampling of antibodies, we have a good estimate of the total number of Covid-19 infections - under 10% with regional variations. And the IFR of around 1% is in the correct ballpark.

It is known that a proportion of people's antibodies fade and that some people don't generate antibodies. Hence any calculation that uses the presence of antibodies as the only test of infection will inevitably have too small a figure. Hence that could only produce an upper bound of the actual figure for Infections Fatality Rate.
.

I'm aware that its an imprecise figure - hence my "ballpark" description. But I think the major reason to hope that on going forward it will reduce, is due to the improvement in treatments, and the reluctance of oldies to catch the infection - by stoutly maintaining social distancing.
Its interesting to see different opinions on risk aversion. I suspect few would frequent a restaurant where 1% of its diners die after eating their meal. But if the Covid-19 infection is allowed to multiply out of control, an oldie with a medical condition may be looking at such a choice.

[dealtn]

Its interesting to see different opinions on risk aversion. I suspect few would frequent a restaurant where 1% of its diners die after eating their meal. But if the Covid-19 infection is allowed to multiply out of control, an oldie with a medical condition may be looking at such a choice.

Well to use the analogy I suspect that many would if the alternative wasn't eating. Clearly that's not the case, there are numerous lower risk ways of sourcing food as an alternative.

Now consider instead of limiting the thought experiment to restaurant use, but the wider category of "living".

If 1% have a certainty of dying from Covid infection (which assumes everyone gets it, and no inoculation via vaccine) but there is no certainty on timing what do you do? At one end of the spectrum is the "don't go out, wrap in cotton wool, delay as long as possible the inevitable", the other is "lead a normal life". (I guess there exists a more extreme spectrum end of search it out and get it over with, one way or the other - but let's dismiss this in the thought experiment).

If the first of these alternatives equates to "no life", and miserable quality of life but lowest risk (longest expected longevity), and the second is reasonable quality of life, yet uncertain longevity (and risk of much shortened longevity). What do we think the projected decision process looks like?

I know my parents, with perhaps life expectancy < 10 years, are closer to the second of these choices.

Even knowing that your choice may have an effect on others might not be sufficient reason to cause you to change course. I guess that's where compulsion (and compliance) work, but such compulsion (and compliance) diminish over time.

[johnhemming]

I'm aware that its an imprecise figure - hence my "ballpark" description.

It is not that it is an imprecise figure, but it is known to be too high. Hence the one thing that is known about it is that it is not in the ballpark.

[UncleEbenezer]

If the first of these alternatives equates to "no life", and miserable quality of life but lowest risk (longest expected longevity), and the second is reasonable quality of life, yet uncertain longevity (and risk of much shortened longevity). What do we think the projected decision process looks like?

I, like most of us, make that decision every day when I step out of the front door and onto a pavement where I'm unprotected if one of the tonnes of metal passing at deadly speed were to go out of control in my direction.

People making that decision not for themselves but for others often make a different choice - as when parents won't let their children go somewhere unaccompanied.

[tjh290633]

If 1% have a certainty of dying from Covid infection (which assumes everyone gets it, and no inoculation via vaccine) but there is no certainty on timing what do you do? At one end of the spectrum is the "don't go out, wrap in cotton wool, delay as long as possible the inevitable", the other is "lead a normal life". (I guess there exists a more extreme spectrum end of search it out and get it over with, one way or the other - but let's dismiss this in the thought experiment).

If the first of these alternatives equates to "no life", and miserable quality of life but lowest risk (longest expected longevity), and the second is reasonable quality of life, yet uncertain longevity (and risk of much shortened longevity). What do we think the projected decision process looks like?

I know my parents, with perhaps life expectancy < 10 years, are closer to the second of these choices.

As one of those, well past my sell-by date, I intend to lead a normal life, taking sensible precautions.

I went to the paper shop this morning, by car as we were having a heavy shower at the time. Donned my mask before entering, waiting for the odd person to leave first. Floundered around in the fog, but I know where my papers will be. Paid with the vouchers and left. Later started on a bramble exorcism exercise, but found a lot more when I tried to remove some tall-growing ones, so just cut them off at head height. Result, 6 bags to go to the dump. Refuelled the wife's car on the way home.

Tomorrow we hope to visit a shielding friend in her garden, weather permitting, then we have a lunch booked at one of my favourite pubs. Social distancing will, of course, be observed.

Thursday will be curry at Wetherspoons.

Friday I have a blood test at the Health Centre.

Saturday will be the weekly shop at Tesco.

That is pretty normal, except for missing lunches with friends.

TJH

[scotia]

I'm aware that its an imprecise figure - hence my "ballpark" description.

It is not that it is an imprecise figure, but it is known to be too high. Hence the one thing that is known about it is that it is not in the ballpark.

Sorry - I'm puzzled by your statement. The recently published English statistics, which (hopefully) we both have read, state the following
"The estimated community IFR (excluding care homes) was 0.9% (0.86, 0.94), increasing with age and
higher in males (1.07%, 1.00, 1.15) than females (0.71%, 0.67, 0.75) (Table 3). Sensitivity analyses
indicate an IFR as high as 1.58% (1.51%, 1.65%) if excess rather than COVID-specific deaths are used
and care home deaths are included"
Did I miss a statement that said that these numbers were known to be too high? Or is it just your opinion?
Personally I think my published statement "And the IFR of around 1% is in the correct ballpark" fits the above figures

[johnhemming]

Did I miss a statement that said that these numbers were known to be too high?

You missed a statement in the same report that said they were making assumptions that could result in the figure being high. I quoted from the report in this thread.

[scotia]

Its interesting to see different opinions on risk aversion. I suspect few would frequent a restaurant where 1% of its diners die after eating their meal. But if the Covid-19 infection is allowed to multiply out of control, an oldie with a medical condition may be looking at such a choice.

Now consider instead of limiting the thought experiment to restaurant use, but the wider category of "living".

I'm not disagreeing with much of what you say - life is for living, and currently myself and my wife (mid seventies) are off enjoying a holiday in the Scottish Highlands.
But I just wanted to pose an easy question which I expect most people would respond with the answer - "I'm certainly not dining there!"
And then I wanted to propose that this choice could become a reality if we don't stick by restrictions that control the infection rate to its present low level. So here's a set of hypotheses to think about - but please take my numbers with considerable circumspection - I only want to describe a possible "what if". First - If we all returned to our pre-pandemic behaviour, the infection would rapidly increase, and we could have a few percent of the populace who are infective - lets say 2%. Now suppose you are one in about about 50 persons dining, but only about 5 are close enough to be considered as contacts, then we have a 0.1 infective contact. Its more difficult to work out the problem with staff. Lets say only 2 waiters are serving your table, however there are 8 kitchen staff who have contact with the plates, cutlery and food. So I suggest this amounts to 0.2 infective contacts. Now Covd-19 is highly infective - so I'm going to propose that each invective contact has a 50% chance of passing over the infection. This gives us, on very crude numbers a 15% chance of infection. Now off to the grim numbers associated with mortality. If you are 75+ in age (as I am) then you have an 11.6% chance of dying from the infection. So there is a 1.74% chance that you will die from this single dining experience. For those in the 45-64 age group this reduces to a 0.075%, and for those even younger it is near zero.
Ok - we can have a good laugh at the numbers, but I'm trying to make clear, particularly to the younger generation, that its a good idea to accept restrictions that control the infection rate, so that the elderly can have a reasonable quality of life.

[scotia]

Did I miss a statement that said that these numbers were known to be too high?

You missed a statement in the same report that said they were making assumptions that could result in the figure being high. I quoted from the report in this thread.

Yes - "could" be too high - not "known to be"
But to end the bickering - lets look at the numbers, and your interpretation of them that leads to your earlier pronouncement that my "around 1%" is definitely out of the ballpark on the high side. Lets start with the 1.58% figure. What adjustments are you going to make to reduce it to definitely below the "around 1%" figure?

[johnhemming]

The difficulty in calculating IFR is knowing the denominator. The "official report" knowingly overstated the IFR and does not reveal any sensitivity calculations on infections without antibodies. Frankly that makes it tendentious.