COVID-19

Discussions about serious topics, for serious people
Locked
Millennie Al
After Pie
Posts: 1621
Joined: Mon Mar 16, 2020 4:02 am

Re: COVID-19

Post by Millennie Al » Mon Mar 01, 2021 2:50 am

Herainestold wrote:
Sun Feb 28, 2021 9:07 pm
Different groups have different susceptibilities and different responses to the virus. Quite large differences in fact.
In treating these different groups you might have to use different strategies and techniques and treatments. So would different vaccines taking into account the differences between groups not make a difference?
Anything is possible in biology, as the systems are so complex, but it's quite reasonable to assume the same vaccine works best for all groups.

It's a bit like dealing with shoplifters. Different shops might be affected in different ways by a thief, but one defence is to distribute a photo of a thief so shopkeepers can recognise them. The same photo works for all shops. Drugs frequently interact with your body to produce an effect, but vaccines are just specimens of something that should be attacked. While one person's immune system may be much better able to attack than another's, that doesn't mean that a different specimen would change that.

However, there are some grounds which might support different vaccines for different people due to the nature of the ones developed for this disease. Vaccines used to be merely weakened or killed versions of the real pathogens, but now we have vaccines which contain only part of the virus (the Oxford one - which is therefore like a photo of the thief's face rather than a full-body shot) and ones which don't contain the virus themselves but cause our bodies to make the relevant part of the virus (Moderna). The latter might well have variable efficacy across different groups, but it's quite reasonable to presume it doesn't until we have evidence that it does.

User avatar
wilsontown
Clardic Fug
Posts: 208
Joined: Tue Feb 23, 2021 11:51 am

Re: COVID-19

Post by wilsontown » Mon Mar 01, 2021 10:24 am

jdc wrote:
Sun Feb 28, 2021 7:23 pm
I think you've missed the context for that question Steamy (see below quote). Wilsontown's trying to find out why we'd need personalised vaccines for different demographics rather than why we need to have more than one manufacturer developing vaccines against coronavirus.
Herainestold wrote:
Sat Feb 27, 2021 10:46 pm
This disease affects groups differently. It is especially bad for POC and BAME people, the elderly, those with hugh BMIs, and strangely enough, males.
Shouldnt there be different vaccines developed for all these subgroups? And how can schools go back when kids aren't vaccinated?
Maybe just different dose vaccines for groups with different risk profiles? Why is nobody asking these questions?
Cheers jdc, that is indeed what I was getting at. SteamTraen's post was interesting anyway, though.
"All models are wrong but some are useful" - George Box

User avatar
Bird on a Fire
Princess POW
Posts: 10137
Joined: Fri Oct 11, 2019 5:05 pm
Location: Portugal

Re: COVID-19

Post by Bird on a Fire » Mon Mar 01, 2021 11:22 am

Do different subgroups have different immune responses to the virus, specifically?

I thought a lot of the differences were due to things like different prevalence of respiratory conditions, behavioural and socioeconomic factors, possibly vitamin D (can't remember if there's a firm conclusion on this yet?) and so on, which aren't really factors a vaccine can address.

I know there have been problems in the past with testing medical treatments on young healthy white men and then expecting them to work the same in other groups, but I don't think this is likely to be one of them - the testing cohorts in places like Brazil are likely to have been pretty diverse.
We have the right to a clean, healthy, sustainable environment.

User avatar
sTeamTraen
After Pie
Posts: 2554
Joined: Mon Nov 11, 2019 4:24 pm
Location: Palma de Mallorca, Spain

Re: COVID-19

Post by sTeamTraen » Mon Mar 01, 2021 5:01 pm

Bird on a Fire wrote:
Mon Mar 01, 2021 11:22 am
Do different subgroups have different immune responses to the virus, specifically?
Until we know more about the subtleties in outcome differences (e.g., do people of recent African ancestry get more liver problems with their long Covid whereas South Asians get more lung damage or whatever), I don't think we can say very much about that. At the moment we mostly have data on death, which is certainly the outcome that focuses people's minds, but is in a sense a one-size-fits-all thing.

Certainly given that death within age groups seems to be strongly correlated with general health, it's going to be hard to say whether there is anything genetic, once you've eliminated the obvious confounds such as socioeconomic status. There really isn't very much difference in the genetics or immune systems of black and white people.

I still haven't been able to understand exactly why France and Germany are recommending against giving the AZ vaccine to over-65s. It could be that it performs less well, but it could also be a combination of smaller samples in the trials combined with a stupid error by a German media outlet (they confused the sample size for the number of people who dies in the trial), which hasn't been completely flushed out of the system. I don't think there's any a priori reason why a conventional vaccine like AZ would be expected to drop off in efficacy in older people.
Something something hammer something something nail

User avatar
wilsontown
Clardic Fug
Posts: 208
Joined: Tue Feb 23, 2021 11:51 am

Re: COVID-19

Post by wilsontown » Mon Mar 01, 2021 5:43 pm

There seems to be reasonable news from today's UK stats as the apparent slow down in the rate of case decline we were all worrying about looks like it might have been a blip after all. The updated graph of % change in 7 day average case numbers from the previous day shows cases continuing to fall strongly. Clearly there's going to be a lot of variability in this particular visualization of the data.
7day_case_rate_change.png
7day_case_rate_change.png (38.07 KiB) Viewed 2816 times
Maybe a bit easier to see if we just look at the last 60 days:
7day_case_rate_change_last60.png
7day_case_rate_change_last60.png (29.8 KiB) Viewed 2816 times
"All models are wrong but some are useful" - George Box

User avatar
lpm
Junior Mod
Posts: 5944
Joined: Mon Nov 11, 2019 1:05 pm

Re: COVID-19

Post by lpm » Mon Mar 01, 2021 6:10 pm

Yep. A sharp dip has now taken the James Annan red 1-week line back below the blue 2-week line.

The smoke alarm detected something, but it probably wasn't smoke. Still, you want an over-sensitive smoke alarm.
⭐ Awarded gold star 4 November 2021

User avatar
Grumble
Light of Blast
Posts: 4747
Joined: Mon Nov 11, 2019 1:03 pm

Re: COVID-19

Post by Grumble » Mon Mar 01, 2021 6:13 pm

Is it possible that this is the vaccine effect showing up already? Or still too early?
where once I used to scintillate
now I sin till ten past three

User avatar
lpm
Junior Mod
Posts: 5944
Joined: Mon Nov 11, 2019 1:05 pm

Re: COVID-19

Post by lpm » Mon Mar 01, 2021 6:19 pm

Should be visible but only for another week or so. Should probably be more visible than it is.

School reopenings next Monday will change the lockdown level - hence the case number a few days after reopening will be impacted by vaccines down and schools up, and it'll be very hard to split out the effect.

If there's a vaccine effect then the one-week average red line should be below the other averages this time next week.
⭐ Awarded gold star 4 November 2021

User avatar
lpm
Junior Mod
Posts: 5944
Joined: Mon Nov 11, 2019 1:05 pm

Re: COVID-19

Post by lpm » Mon Mar 01, 2021 6:47 pm

Woah, this has shifted fast.

Here's the James Annan chart a week ago. The red line above the others, showing a slow down in the improvement.

Followed by the chart he's just updated for today's figures. A reversal - cases in the past week now declining faster.

The vaccine effect is one possible explanation.

Image

Image
⭐ Awarded gold star 4 November 2021

Herainestold
After Pie
Posts: 2029
Joined: Mon Nov 25, 2019 1:23 pm

Re: COVID-19

Post by Herainestold » Mon Mar 01, 2021 7:55 pm

lpm wrote:
Mon Mar 01, 2021 6:47 pm
Woah, this has shifted fast.

Here's the James Annan chart a week ago. The red line above the others, showing a slow down in the improvement.

Followed by the chart he's just updated for today's figures. A reversal - cases in the past week now declining faster.

The vaccine effect is one possible explanation.

Image

Image
This is indeed good news, but we should be cautious about broadcasting it. We do really need to stay locked down for at least 9 more weeks.
Masking forever
Putin is a monster.
Russian socialism will rise again

User avatar
Bird on a Fire
Princess POW
Posts: 10137
Joined: Fri Oct 11, 2019 5:05 pm
Location: Portugal

Re: COVID-19

Post by Bird on a Fire » Mon Mar 01, 2021 8:00 pm

I worry that we might be venturing into this territory:

Image

How far outside Annan's credible intervals is the recent wobble?
We have the right to a clean, healthy, sustainable environment.

User avatar
lpm
Junior Mod
Posts: 5944
Joined: Mon Nov 11, 2019 1:05 pm

Re: COVID-19

Post by lpm » Mon Mar 01, 2021 8:16 pm

Bird on a Fire wrote:
Mon Mar 01, 2021 8:00 pm
I worry that we might be venturing into this territory:

How far outside Annan's credible intervals is the recent wobble?
That is not what it's for. It's just an immediate and easily understood picture.

This is the exam question all over again. Your smoke alarm doesn't need to bother with credible intervals, it just needs to sound an alarm the second it's the slightest bit worried. An immediate investigation follows. You don't want an alarm that studies the data over 7 minutes before determining the particles have exceeded standard norms.

"Simple" can often defeat "proper" - for example our simple spreadsheets one year ago were far better than the disastrous models used by the govt. "Four weeks behind Italy". The proper models were 8x worse than ours.
⭐ Awarded gold star 4 November 2021

User avatar
Bird on a Fire
Princess POW
Posts: 10137
Joined: Fri Oct 11, 2019 5:05 pm
Location: Portugal

Re: COVID-19

Post by Bird on a Fire » Mon Mar 01, 2021 8:32 pm

Sure - I can see the advantage of an early warning system that's a bit over-sensitive.

But in this case it went up a bit and then down a bit again, with nobody actually taking any action in the meantime (unless I missed something important).

So we've passed the need for an alarm and are now able to try to understand things properly - did anything actually happen, or was it just a random bit of wobbling?
We have the right to a clean, healthy, sustainable environment.

User avatar
lpm
Junior Mod
Posts: 5944
Joined: Mon Nov 11, 2019 1:05 pm

Re: COVID-19

Post by lpm » Mon Mar 01, 2021 8:39 pm

https://twitter.com/jamesannan for the other charts - which have proper ranges - and links to the full model.

But the answer is that no-one will ever know.
⭐ Awarded gold star 4 November 2021

User avatar
Bird on a Fire
Princess POW
Posts: 10137
Joined: Fri Oct 11, 2019 5:05 pm
Location: Portugal

Re: COVID-19

Post by Bird on a Fire » Mon Mar 01, 2021 8:52 pm

Well, the slightly higher number of deaths for a few days was still within the range his model predicted, so there's no real reason to go hunting for an explanation.

Same kind of thing as confusing weather and climate. A few cold winters in a row don't disprove climate change, and a few extra deaths don't mean the UK is deviating from an exponential trajectory.
We have the right to a clean, healthy, sustainable environment.

User avatar
discovolante
Stummy Beige
Posts: 4084
Joined: Fri Oct 11, 2019 5:10 pm

Re: COVID-19

Post by discovolante » Mon Mar 01, 2021 9:17 pm

Herainestold wrote:
Mon Mar 01, 2021 7:55 pm
lpm wrote:
Mon Mar 01, 2021 6:47 pm
Woah, this has shifted fast.

Here's the James Annan chart a week ago. The red line above the others, showing a slow down in the improvement.

Followed by the chart he's just updated for today's figures. A reversal - cases in the past week now declining faster.

The vaccine effect is one possible explanation.

Image

Image
This is indeed good news, but we should be cautious about broadcasting it. We do really need to stay locked down for at least 9 more weeks.
On the other hand, it shows a potential endpoint and some reassurance that we won't need to be locked down 'forever', so it's not right to say that we may as well just open up now.
To defy the laws of tradition is a crusade only of the brave.

KAJ
Fuzzable
Posts: 310
Joined: Thu Nov 14, 2019 5:05 pm
Location: UK

Re: COVID-19

Post by KAJ » Mon Mar 01, 2021 10:51 pm

After our last exchange I gave some thought to what James Annan seems to be doing. I don't do or really understand that Twitter thing. I tried following the link in ...
lpm wrote:
Mon Mar 01, 2021 8:39 pm
https://twitter.com/jamesannan for the other charts - which have proper ranges - and links to the full model.
... but couldn't easily see any explanation, so I've tried to deduce it from the charts.

Differencing a time series is a quite conventional way to elucidate rate of change rather than value; taking ratios is equivalent to differencing logs. Using a 7-day lag is a convenient way to minimise any day-of-week effect. Differences do tend to emphasise "noise" and moving averages are a quite conventional way to minimise this. An alternative would be to moving average the original values and difference the averages, I haven't investigated that (CBA).

Plotting the final moving average as a horizontal line with length equal to the number of averaged points is unusual, and I can't see any good reason. It is more common to plot each average as a single point, whether in the middle of the averaged X range or at one end - I tend to plot at the last averaged X. This means you can plot the 7-day (e.g.) moving average for the period ending today, or yesterday, or the day before, ... I tend to plot these as a line. This makes clear whether and how fast the average is increasing or decreasing. There is no need to plot a longer period average with which to compare a shorter period to discern whether the latter is changing - that seems very strange. Nevertheless, I've plotted the 2 and 3 week moving averages as well. This is what I get.
diffPubCases.png
diffPubCases.png (34.46 KiB) Viewed 2687 times

User avatar
Bird on a Fire
Princess POW
Posts: 10137
Joined: Fri Oct 11, 2019 5:05 pm
Location: Portugal

Re: COVID-19

Post by Bird on a Fire » Mon Mar 01, 2021 11:11 pm

Ah, I think he's not plotting 7- vs 14- (etc-)day rolling averages, but the average of the most recent 7 vs 14 etc days - ie, averaging over the period shown by the lines.

All his stuff is quite unconventional, though. He models the pandemic trajectory as a random walk with no underlying statistical model whatsoever, just fitting to the data.

His plots do seem to do a good job of fitting, but I can't recall seeing a comparison of say, last month's forecast with the actual observed figures, because the plots he posts each day are based on an updated version of the model.
We have the right to a clean, healthy, sustainable environment.

User avatar
Bird on a Fire
Princess POW
Posts: 10137
Joined: Fri Oct 11, 2019 5:05 pm
Location: Portugal

Re: COVID-19

Post by Bird on a Fire » Mon Mar 01, 2021 11:18 pm

sTeamTraen wrote:
Mon Mar 01, 2021 5:01 pm
Bird on a Fire wrote:
Mon Mar 01, 2021 11:22 am
Do different subgroups have different immune responses to the virus, specifically?
Until we know more about the subtleties in outcome differences (e.g., do people of recent African ancestry get more liver problems with their long Covid whereas South Asians get more lung damage or whatever), I don't think we can say very much about that. At the moment we mostly have data on death, which is certainly the outcome that focuses people's minds, but is in a sense a one-size-fits-all thing.

Certainly given that death within age groups seems to be strongly correlated with general health, it's going to be hard to say whether there is anything genetic, once you've eliminated the obvious confounds such as socioeconomic status. There really isn't very much difference in the genetics or immune systems of black and white people.

I still haven't been able to understand exactly why France and Germany are recommending against giving the AZ vaccine to over-65s. It could be that it performs less well, but it could also be a combination of smaller samples in the trials combined with a stupid error by a German media outlet (they confused the sample size for the number of people who dies in the trial), which hasn't been completely flushed out of the system. I don't think there's any a priori reason why a conventional vaccine like AZ would be expected to drop off in efficacy in older people.
Thanks for this, btw - more or less what I thought.

And I'm not sure there's much point in designing, say, a special vaccine for "black people" (containing a huge amount of genetic variation that's not well correlated with skin colour) when uptake of the vaccines we already have are lower in black communities. Makes more sense to focus on getting trusted advice to those people than start a load of complex community immunological assays.
We have the right to a clean, healthy, sustainable environment.

User avatar
lpm
Junior Mod
Posts: 5944
Joined: Mon Nov 11, 2019 1:05 pm

Re: COVID-19

Post by lpm » Mon Mar 01, 2021 11:22 pm

KAJ wrote:
Mon Mar 01, 2021 10:51 pm
Nevertheless, I've plotted the 2 and 3 week moving averages as well. This is what I get.
diffPubCases.png
That's a good chart.

It shows a pretty stable picture for most of the lockdown. The rise in the red line is then pretty exceptional. Definitely enough to sound an alarm.

We need to be nimble in the coming weeks. The govt is leaving 5 weeks between unlockdown steps. Which isn't long - due to the lags. If schools reopen on 8th March and cause a rise in infections by 15th, then by the time people notice symptoms, get tested and the numbers are collated it's the 22nd, and then we track those numbers in a 7 day average with a noticeable effect by 29th, and we watch it till 5th April to be sure - and then it's too late to postpone the next unlockdown on 12th when businesses have already made their plans.

If things get dangerous, only a fast and responsive dashboard will warn this reckless govt to brake in time.

Which is why I think patient, proper models are the wrong approach - a quick and dirty method can have more value.
⭐ Awarded gold star 4 November 2021

User avatar
sTeamTraen
After Pie
Posts: 2554
Joined: Mon Nov 11, 2019 4:24 pm
Location: Palma de Mallorca, Spain

Re: COVID-19

Post by sTeamTraen » Tue Mar 02, 2021 12:08 am

Bird on a Fire wrote:
Mon Mar 01, 2021 11:18 pm
And I'm not sure there's much point in designing, say, a special vaccine for "black people" (containing a huge amount of genetic variation that's not well correlated with skin colour) when uptake of the vaccines we already have are lower in black communities. Makes more sense to focus on getting trusted advice to those people than start a load of complex community immunological assays.
I'm also not sure how a person who may be suspicious of the vaccine in the first place would react to a "special version just for you", developed by the same people from other ethnic groups who made the original. Indeed, something like this attitude is probably at work in this example, in a country where many people like to think they are in the only ethnic group:
Herainestold wrote:
Fri Feb 26, 2021 1:05 am
Hungary is administering five vaccines. Pfizer, Moderna, AZ, Sputnik and Sinopharm
A survey of 1,000 people in the capital of Budapest by pollster Median and the 21 Research Center showed that among those willing to be vaccinated, only 27% would take a Chinese vaccine and 43% a Russian vaccine, compared to 84% who would take a jab developed in Western countries. The poll, which was conducted at the end of January, had a margin of error of plus or minus 3%.
Something something hammer something something nail

Herainestold
After Pie
Posts: 2029
Joined: Mon Nov 25, 2019 1:23 pm

Re: COVID-19

Post by Herainestold » Tue Mar 02, 2021 2:17 am

sTeamTraen wrote:
Tue Mar 02, 2021 12:08 am
Bird on a Fire wrote:
Mon Mar 01, 2021 11:18 pm
And I'm not sure there's much point in designing, say, a special vaccine for "black people" (containing a huge amount of genetic variation that's not well correlated with skin colour) when uptake of the vaccines we already have are lower in black communities. Makes more sense to focus on getting trusted advice to those people than start a load of complex community immunological assays.
I'm also not sure how a person who may be suspicious of the vaccine in the first place would react to a "special version just for you", developed by the same people from other ethnic groups who made the original. Indeed, something like this attitude is probably at work in this example, in a country where many people like to think they are in the only ethnic group:
Herainestold wrote:
Fri Feb 26, 2021 1:05 am
Hungary is administering five vaccines. Pfizer, Moderna, AZ, Sputnik and Sinopharm
A survey of 1,000 people in the capital of Budapest by pollster Median and the 21 Research Center showed that among those willing to be vaccinated, only 27% would take a Chinese vaccine and 43% a Russian vaccine, compared to 84% who would take a jab developed in Western countries. The poll, which was conducted at the end of January, had a margin of error of plus or minus 3%.
The solution might to have a Black team develop a vaccine for people of African descent. I recall an American BLM activist demanding that African American covid patients be treated in a Black hospital by Black staff. Oregon state are vaccinating BAME and POC people as a priority, I dont know what the uptake is like, if it is less than the average nor not.
Vaccines from China -there are three now, I think -and from the Serum institute in India, could be imported for own Asian population.

Going forward we are going to have to think about health care development and delivery from an equity point of view, which has been neglected up to now.
Masking forever
Putin is a monster.
Russian socialism will rise again

User avatar
Martin_B
After Pie
Posts: 1614
Joined: Mon Nov 11, 2019 1:20 pm
Location: Perth, WA

Re: COVID-19

Post by Martin_B » Tue Mar 02, 2021 3:21 am

Herainestold wrote:
Tue Mar 02, 2021 2:17 am
The solution might to have a Black team develop a vaccine for people of African descent. I recall an American BLM activist demanding that African American covid patients be treated in a Black hospital by Black staff. Oregon state are vaccinating BAME and POC people as a priority, I dont know what the uptake is like, if it is less than the average nor not.
Vaccines from China -there are three now, I think -and from the Serum institute in India, could be imported for own Asian population.

Going forward we are going to have to think about health care development and delivery from an equity point of view, which has been neglected up to now.
What makes you think that the team of black researchers would develop the optimum vaccine for people of African descent? Are they going to identify vaccines which work well for non-African people and abandon that line of research because they're concentrating on people of African descent only?

Are you sure that the three Chinese vaccines are the best vaccines for Asian populations?
If the AstraZeneca vaccine turns out to proffer an advantage for, say, Asian people, should British Asians (or even the general Asian population) use a Chinese- or Indian-made vaccine, if the AZ vaccine is cheaper and available?

Yes, there may be (probably are) problems with a lack of certain minorities in medical research, but is the creation of single-minority-only research teams concentrating on their single-minority-only issues really the optimum way forward?

There are issues with drug companies mainly performing research in the rich companies, and historically drug trials in 3rd world countries being performed on the cheap, or without fully informed consent, and I think that more drug development in poorer countries can only be a good thing, but drug development costs a lot of money and poorer countries usually consider that they have better economic priorities than drug R&D.
"My interest is in the future, because I'm going to spend the rest of my life there"

tom p
After Pie
Posts: 1876
Joined: Mon Nov 11, 2019 1:14 pm
Location: the low countries

Re: COVID-19

Post by tom p » Tue Mar 02, 2021 3:57 pm

Herainestold wrote:
Tue Mar 02, 2021 2:17 am
The solution might to have a Black team develop a vaccine for people of African descent. I recall an American BLM activist demanding that African American covid patients be treated in a Black hospital by Black staff. Oregon state are vaccinating BAME and POC people as a priority, I dont know what the uptake is like, if it is less than the average nor not.
Vaccines from China -there are three now, I think -and from the Serum institute in India, could be imported for own Asian population.

Going forward we are going to have to think about health care development and delivery from an equity point of view, which has been neglected up to now.
This is so dumb, it's not even wrong. It's too wrong to be categorisable.
You don't have a clue what you're talking about and you're asking the wrong questions in a ridiculous way.
It's time you hushed now to stop yourself looking any sillier.

tom p
After Pie
Posts: 1876
Joined: Mon Nov 11, 2019 1:14 pm
Location: the low countries

Re: COVID-19

Post by tom p » Tue Mar 02, 2021 3:58 pm

sTeamTraen wrote:
Mon Mar 01, 2021 5:01 pm
I still haven't been able to understand exactly why France and Germany are recommending against giving the AZ vaccine to over-65s. It could be that it performs less well, but it could also be a combination of smaller samples in the trials combined with a stupid error by a German media outlet (they confused the sample size for the number of people who dies in the trial), which hasn't been completely flushed out of the system. I don't think there's any a priori reason why a conventional vaccine like AZ would be expected to drop off in efficacy in older people.
It's the bold bit, coupled with attention-seeking politicians feeling they need to say something when keeping their big mouths shut would have been better.

Locked