Covid-19 #25: The Prisoner’s Dilemma

[Ormond]

In terms of vaccine "efficacy" -- as I understand it vaccine experts define "efficacy" as the % of vaccinated people who end up with no symptoms at all.  Most vaccines do result in those who develop symptoms having milder cases on the average. With a disease as deadly as Covid-19, a 50% "effective" vaccine would still be well worth it, especially in countries with less developed healthcare, if it drastically lowers the hospitalization and death rate in the 50% of persons for whom it is technically "ineffective." And as I understand it, that is indeed the case with the Covid-19 vaccines: the few who do still get sick are much more likely to just have mild common-cold-like symptoms than the unvaccinated are. 

[DireWolfSpirit]

Apologies if this has already been posted, good news on the Pfizer vaccine storage front.

https://www.foxbusiness.com/politics/pfizer-says-its-covid-vaccine-doesnt-need-to-be-stored-at-ultra-freezing-temps-anymore

[Fragile Bird]

14 minutes ago, DireWolfSpirit said:

Apologies if this has already been posted, good news on the Pfizer vaccine storage front.

https://www.foxbusiness.com/politics/pfizer-says-its-covid-vaccine-doesnt-need-to-be-stored-at-ultra-freezing-temps-anymore

It's actually what we've known for some time, but vaccine stability tests require months of data.

[Mlle. Zabzie]

4 minutes ago, Fragile Bird said:

It's actually what we've known for some time, but vaccine stability tests require months of data.

Still great news. 

[Mudguard]

10 hours ago, ljkeane said:

Given the nature of the study I wouldn't assume the numbers on efficacy are exactly right (I don't think it really matters) but are they obviously contradictory? I wouldn't know but I'd assume the people who wrote the paper probably would and they seem fine with it.

I'm not sure what the BBC are supposed to do though. They're accurately reporting what the research paper says. I suppose they could qualify it by saying there's quite a large confidence interval involved but I don't think it really makes a difference to the relevant points of the paper.

Well, the efficacy numbers matter because governments appear to be using them to justify spacing the dosing out to 3 months.  Because the confidence intervals are so large, it's possible that the efficacy is actually worse when dosing is delayed to 3 months.  The original study was not designed to determine whether spacing the dosing less than 6 weeks or over 12 weeks was better.  In fact, the dosing schedule for the trial was supposed to be two standard doses 29 days apart.  User error resulted in lower doses being given to some patients, and problems with vaccine supply resulted in the spaced out dosing exceeding the original 29 day spacing.  It's unbelievable to me that AstraZeneca was so late in providing doses for the clinical trial after it started that doses were given two months or more late.  Both these things happening suggest that the clinical trial was very sloppily run, and don't actually inspire confidence.  US trial is wrapping up soon, and hopefully it was better run.

The short BBC article should have also made it clear that the study was not peer reviewed, if it was based on the paper you linked to, and that the authors were not independent researchers but instead were tied to Oxford and/or AstraZeneca.  This same data was interpreted early on as suggesting that the lower dosing followed by the standard dosing resulted in a higher efficacy than two standard doses, but now, they are saying that they think increasing the spacing between dosings results in higher efficacy.  Maybe, but they really need to do another study if they want the evidence to be scientifically rigorous.  Right now, the data is just suggestive.

One of the findings reported in the BBC article and the paper that appears contradictory is that the single dose efficacy is supposed to be around 75% between 22 and 90 days, but apparently if you give the second dose by 6 weeks or earlier, the efficacy is only 55%.  This finding is counterintuitive because you wouldn't expect the efficacy to go down after receiving the booster shot.  Normally, it should go up.  Note also that the original dosing schedule was supposed to be 29 days, based on the antibody titer results from the earlier phase i/ii studies.  Were they just really lucky that they fucked up the clinical trial so badly and ended up discovering a better dosing schedule?  Maybe.  But maybe the efficacy numbers don't hold up in a properly designed study.

AstraZenenca is doing a lot to back the UK government, from prioritizing doses to promoting studies that back the UK's rollout strategy.  In normal circumstances, I don't think any government agency would recommend using the alternative dosing schedule based on selectively slicing and dicing the data from a study that was never intended to analyze the dosing schedule.  It's a combination of desperation and politics that are resulting in these types of recommendations that aren't really backed yet by the science.  The vaccines were approved and rolled out at the height of the second wave when things were getting really bad, and I think that some degree of panic played a role in the recommendations.  I don't think the rollout with the spaced dosing will be disastrous, but let's just recognize that the evidence backing it is actually pretty flimsy right now.

[Mr. Chatywin et al.]

7 hours ago, Winterfell is Burning said:

The Health Minister of Argentina has resigned after it was revealed he set up a VIP vaccination scheme, to give powerful people who wouldn't be in eligible access to early vaccinations.

So disappointing, but entirely unshocking. 

[ljkeane]

4 minutes ago, Mudguard said:

Well, the efficacy numbers matter because governments appear to be using them to justify spacing the dosing out to 3 months.  Because the confidence intervals are so large, it's possible that the efficacy is actually worse when dosing is delayed to 3 months. 

It's extremely unlikely from what I can tell. The overlap in confidence interval is towards the higher end for the 6 weeks or less anyway and the underlying immune response results were consistent with a far stronger response to a longer gap between doses. Basically it's very likely better to have a 12 week gap and at very worst only slightly worse and still within acceptable levels on the downside.

12 minutes ago, Mudguard said:

The original study was not designed to determine whether spacing the dosing less than 6 weeks or over 12 weeks was better.  In fact, the dosing schedule for the trial was supposed to be two standard doses 29 days apart.  User error resulted in lower doses being given to some patients, and problems with vaccine supply resulted in the spaced out dosing exceeding the original 29 day spacing.

Well actually the original study was done on the basis of it being a single dose vaccine thus a lot of this data being available. But, whatever, they aren't claiming it was a study designed to look at the gap between doses. They're just presenting the available information on the subject.

19 minutes ago, Mudguard said:

One of the findings reported in the BBC article and the paper that appears contradictory is that the single dose efficacy is supposed to be around 75% between 22 and 90 days, but apparently if you give the second dose by 6 weeks or earlier, the efficacy is only 55%.  This finding is counterintuitive because you wouldn't expect the efficacy to go down after receiving the booster shot.  Normally, it should go up.

They aren't comparing the same thing. It's comparing the efficacy between 22 and 90 days of a single dose to the overall efficacy of two doses with less than a six week gap. If the overall efficacy of a single dose was higher than two doses then that would raise some questions about but that's not what it's saying.

[Winterfell is Burning]

7 hours ago, rotting sea cow said:

This is beyond scummy.

And worrying for those of us that live in Brazil and had elder relatives vaccinating- since I wasn't present, I don't know for sure that didn't happen.

And of course there's the risk that a few months from now, some of these people that got air vaccines and didn't realize die from COVID, people will begin to use that as proof that they don't work.

16 minutes ago, Tywin et al. said:

So disappointing, but entirely unshocking. 

I'm pretty sure this is happening in a lot of places, it's just that people haven't been caught- or in authoritarian regimes, it was caught, but suppressed. 

[ljkeane]

35 minutes ago, Mudguard said:

The short BBC article should have also made it clear that the study was not peer reviewed, if it was based on the paper you linked to, and that the authors were not independent researchers but instead were tied to Oxford and/or AstraZeneca. 

It has been now by the way. It was first reported on back when that pre print was made available for peer review. The reason they've mentioned it again is apparently because it was formally published a few days ago.

[Fragile Bird]

1 hour ago, Mudguard said:

 Both these things happening suggest that the clinical trial was very sloppily run, and don't actually inspire confidence.  US trial is wrapping up soon, and hopefully it was better run.

At the risk of repeating myself, I said the US government wasn’t going to approve the AZ vaccine off of the trial the UK government used. Lots of doctors interviewed on CNN made the comment at the time. Canada wasn’t going to either. So those of you who were shocked and horrified that Canada indicated it would take delivery of vaccine we are entitled to under COVAX, you can be sure that won’t happen until the US trial is complete and the vaccine is approved.

The president of the Indian vaccine manufacturer tweeted that he had spoken to PM Trudeau and said they could ship us vaccine next month, but let’s see what that Phase 3 trial says first.

[Padraig]

Forbes had an article on US approval a few days ago.  Here.  Reaffirms what Fragile Bird said.

Quote

So, why hasn’t the FDA authorized it yet? The main reason is simple: the company hasn’t applied for emergency-use authorization yet. 

...

The FDA wants to see the U.S. trial that’s ongoing.  A spokesperson for the company said this data would be available in “the coming weeks,” but wouldn’t provide a more concrete time line. 

We may see results in March.  Thus, possible approval in April?  I believe J&J may get approval at the end of the coming week.  If I remember correctly.  Good times.  EMA approval likely in mid-March.  And J&J has also applied for WHO approval, so its full steam ahead.

2 hours ago, ljkeane said:

They aren't comparing the same thing. It's comparing the efficacy between 22 and 90 days of a single dose to the overall efficacy of two doses with less than a six week gap.  If the overall efficacy of a single dose was higher than two doses then that would raise some questions about but that's not what it's saying.

I don't quite get that.   The following seem to be the 2 relevant lines from this paper, with CI for clarity.

Vaccine efficacy after 2 standard doses rose from 54.9% (32.7%, 69.7%) with an interval < 6 weeks, to 82.4% (62.7%, 91.7%) when spaced more than 12 weeks apart (Table 1).

A single standard dose of vaccine provided protection against primary symptomatic COVID-19 in the first 90 days of 76%, (95%CI, 59%, 86%).

So you are saying that they are comparing the efficacy between 22 and 90 days of a single dose (which is 76% based on the above) to the overall efficacy of two doses with less than a six week gap (which is 54.9% from the above).  Thus, the overall efficacy of a single dose is higher than 2 doses (at 6 weeks out)? 

Now, at 12 weeks, 2 doses is higher (82.4% v 76%).  But I don't see why that means we can ignore the original comparison at 6 weeks?

Of course, if you look at the confidence intervals, it does all work out.  That's what I realised yesterday.  But I think you are making a different point?

3 hours ago, ljkeane said:

Well actually the original study was done on the basis of it being a single dose vaccine thus a lot of this data being available.

Huh.  I thought this study was simply examining the efficacy of single doses (and it evolved).  Not that the vaccine itself was supposed to be a single dose originally.  Interesting.

[ljkeane]

6 hours ago, Padraig said:

So you are saying that they are comparing the efficacy between 22 and 90 days of a single dose (which is 76% based on the above) to the overall efficacy of two doses with less than a six week gap (which is 54.9% from the above).  Thus, the overall efficacy of a single dose is higher than 2 doses (at 6 weeks out)? 

They aren’t comparing them at all. They’re doing two separate things. Comparing the overall efficacy between dosing gaps of less than 6 weeks and 12 weeks and then looking if protection from a single dose is maintained during that 12 week gap.

[rotting sea cow]

12 hours ago, Mudguard said:

Well, the efficacy numbers matter because governments appear to be using them to justify spacing the dosing out to 3 months.  Because the confidence intervals are so large, it's possible that the efficacy is actually worse when dosing is delayed to 3 months.  The original study was not designed to determine whether spacing the dosing less than 6 weeks or over 12 weeks was better.  In fact, the dosing schedule for the trial was supposed to be two standard doses 29 days apart.  User error resulted in lower doses being given to some patients, and problems with vaccine supply resulted in the spaced out dosing exceeding the original 29 day spacing.  It's unbelievable to me that AstraZeneca was so late in providing doses for the clinical trial after it started that doses were given two months or more late.  Both these things happening suggest that the clinical trial was very sloppily run, and don't actually inspire confidence.  US trial is wrapping up soon, and hopefully it was better run.

(snip)

AstraZenenca is doing a lot to back the UK government, from prioritizing doses to promoting studies that back the UK's rollout strategy.  In normal circumstances, I don't think any government agency would recommend using the alternative dosing schedule based on selectively slicing and dicing the data from a study that was never intended to analyze the dosing schedule.  It's a combination of desperation and politics that are resulting in these types of recommendations that aren't really backed yet by the science.  The vaccines were approved and rolled out at the height of the second wave when things were getting really bad, and I think that some degree of panic played a role in the recommendations.  I don't think the rollout with the spaced dosing will be disastrous, but let's just recognize that the evidence backing it is actually pretty flimsy right now.

Thanks for your views. Very well written argumentation. Governments keep claiming they are following sound scientific advice, which plainly they do not. Look at this article in The Lancet about the same issue. Pretty damming.

https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(21)00455-4/fulltext

and this is troubling

Quote

Sub-optimal vaccination will create selective pressure facilitating the emergence of vaccine-resistant variants, which could result in a persisting pandemic. New vaccines, covering such variants, can be made but will require time for testing, mass production, and distribution

 

 

[Heartofice]

So today here in the UK Boris is supposedly going to outline our timetable out of lockdown. 
 

Next month we’ll have schools going back , which is great. 
 

Then two people will be able to go to a park and meet and sit down and have a coffee.

What is laughable about that suggestion  is that it’s based on the idea that people aren’t doing it already. 
 

Here’s the rub, I think adherence to the lockdown guidelines has been pretty low when it comes to certain situations, and has been for quite some time. I think most people are kind of tired of restrictions and taking their own chances and deciding what is a worthwhile risk.

Yesterday I went out to a local park and it was busy, lots of different groups out and about. The local town had people milling about. Any rule about two people sitting next to each other has been completely disregarded.

There is still of course the point that most people maintain social distances from strangers, I still see handwashing etc.  But overall I don’t think this is as much of an easing of restrictions as the government thinks 

[BigFatCoward]

I went for a walk yesterday and would estimate at least 50% of people looked (you cant ever be fully sure about household makeup) that they were not abiding by the rules. 

 

[Padraig]

4 hours ago, ljkeane said:

They’re doing two separate things. Comparing the overall efficacy between dosing gaps of less than 6 weeks and 12 weeks and then looking if protection from a single dose is maintained during that 12 week gap.

Right.  When I looked at that paper, it seemed reasonable to assume they knew what they are doing.  Still does.  Even if what you say sounds odd to me.

[Heartofice]

20 minutes ago, BigFatCoward said:

I went for a walk yesterday and would estimate at least 50% of people looked (you cant ever be fully sure about household makeup) that they were not abiding by the rules. 

 

Well yeah that would be my take too. This comes back to just how much genuine effect a government can really have on transmission levels without being completely authoritarian about it. 
 

I’d say most people are now educated enough about the disease to make their own minds up, and so they probably think it’s fine to go outsid in mixed groups with people they know. You could of course have the police running around asking everyone who they are etc, but nobody really wants that either.

[ljkeane]

53 minutes ago, Padraig said:

Right.  When I looked at that paper, it seemed reasonable to assume they knew what they are doing.  Still does.  Even if what you say sounds odd to me.

The reason they decided not to continue with the vaccine as a single dose was because they thought it needed a booster to provide longterm protection. Apparently by 180 days after vaccination with a single dose the antibody response had declined by 64%.

So what you're doing by looking at the efficacy of a single dose between 22 and 90 days is looking at the period when it, in theory, should offer good protection and cutting off the period when it should get significantly worse. That's fine if all you care about is that period but it means it's not a meaningful comparison if you're using it to compare the efficacy of a single dose with the overall efficacy of two doses with whatever dosing gap.

[Which Tyler]

1 hour ago, BigFatCoward said:

I went for a walk yesterday and would estimate at least 50% of people looked (you cant ever be fully sure about household makeup) that they were not abiding by the rules.

I have to say, my experience in semi-rural Gloucestershire is extremely different - yourself and HOI are both London yes?.

You see the odd group who look suspiciously similar aged, but they're few and far between; mostly see quite a lot of couples, usually masked, and pretty much everyone is giving wide berths when passing.
I've yet to see a group stop to chat and NOT keep 2m+ part; and hardly ever see a bench being used.

[Heartofice]

13 minutes ago, Which Tyler said:

I have to say, my experience in semi-rural Gloucestershire is extremely different - yourself and HOI are both London yes?.

You see the odd group who look suspiciously similar aged, but they're few and far between; mostly see quite a lot of couples, usually masked, and pretty much everyone is giving wide berths when passing.
I've yet to see a group stop to chat and NOT keep 2m+ part; and hardly ever see a bench being used.

Yeah might be that different areas might have different experiences and behaviours. 

What I think is that people are sticking to the rules that make sense to them, and being less cautious about ones that seem overly cautious. So pretty much everyone I've seen for almost a year has been respectful of social distancing, because there is an element of bad manners to not stick to that.

But I'm certain that there has been a lot of household mixing over the past few months. I know plenty of people who've been to parties. Without being too much of a curtain twitcher, I know my neighbours have all had friends and relatives over recently. I've seen out and about people are hanging around with others who are probably not from the same household. I think people are taking calculated risks, figuring they either know people enough to think it's fine or that they are young and so it isn't really an issue for them.

Then there is also a bunch of obvious stuff that nobody talks about, like underground hairdressers. There was a fuss the other week when a footballer posted a photo of themselves having a haircut, which is clearly not allowed. But I've yet to see a footballer without a fresh haircut, every week! This stuff is all going on, it's just not talked about.