> My view is that it would be good if the J&J vaccine was followed by a booster–perhaps of some other vaccine–but that it’s individually fine and in fact socially beneficial to get more people protected quickly by delaying the booster for at least 12 weeks to when vaccines are less scarce. I don’t currently see a reason for thinking differently about the Pfizer and Moderna vaccines.
Not an expert, not a researcher, but an economist...
Regardless of his qualifications, his point is interesting: J&J seems to be about as effective as a single dose of Pfizer. There was a whole bunch of worry about what would happen if people got only one of their two Pfizer doses.
Was that worry all for nothing? Was it backed up by any science? If there was legitimate reason for concern, why do we not have the same concern around the j&j vaccine?
(I'm not trying to advocate any of these points. I'm even less qualified than an economist. But they seem like legitimate questions)
> Was that worry all for nothing? Was it backed up by any science? If there was legitimate reason for concern.
I believe it was because of a lack of evidence either way. The Pfizer trials were two doses. So we didn't know for sure how well one would work. The decision to only do one was made on the basis that this works well for most (but not all) vaccines. It is now known that it indeed works for the Pfizer vaccine, and so it seems to have been a good decision. The concern was that it would not and would set the vaccination programme back by months.
> Why do we not have the same concern around the j&j vaccine?
I'm not sure, but I suspect the J&J vaccine was trialed as one dose from the outset.
> Regardless of his qualifications, his point is interesting: J&J seems to be about as effective as a single dose of Pfizer.
The most compelling J&J number is, IMHO, the 100% efficacy at preventing hospitalizations and deaths. Is a single shot of the double-shot vaccines that effective? AFAIK that study was never run..
The trouble with all of these 100% numbers is that very few people (in either group of trial participants) were hospitalised.
We have a small sample for the estimate of this efficacy. That being said, all the vaccines have shown this in small samples, so my current working theory is that the hospitalisation rates post vaccine will be very, very low (but almost certainly not 0%).
This is more about looking at the ability of our systems to change their response in emergencies; why are obvious things so hard to do, when they both fully make sense, and are supported by an interested group of experts, against a larger but less interested group of conventional, rule-following experts?
What is the relationship of his credentials to his logical argument? By arguing from credentials you implicitly take the side of the status quo in history; you'd have been arguing against every overturning of conventional wisdom.
I'm not saying all rejection of convention is correct, obviously - but if you look at the specifics of the argument, not the status or credentials of the arguer, it seems more likely to come to a correct judgement.
Oh god that first comment rings so depressingly true. Is this just the nature of mankind forever or is there some way of setting up incentives such that this doesn't play out exactly the same way every time?
Yes but AFAIK we don't do it. Incentive structures can be treated as causal structures if you simplify by saying that incentives "cause" people to act a certain way. The disciplines involved in fixing or improving causal structures are systems theory and cybernetics. But these tools are extremely under utilised.
Michael Osterholm, Regents Professor, McKnight Presidential Endowed Chair in Public Health, the director of the Center for Infectious Disease Research and Policy (CIDRAP) and state epidemiologist for the Minnesota Department of Health:
…Imagine you are setting across the table from two people both of whom are 65 or older, both with underlying health conditions. You have two doses of vaccine, one in each hand. And you say to them I can give two doses to you or to you but then the other person gets nothing. Or I can give one dose to both of you. And this is what I know. At the very least, one dose is likely to prevent serious illness, hospitalization and death. Two doses will probably even prevent clinical disease with B.1.1.7. But the other one of you; if you get infected with this virus, which I think substantial numbers of Americans will, things are not looking good for you. What do you want me to do?
If that is your Mom or Dad. Your Grandpa or Grandma. What would you do?
This is where the rubber meets the road. I think if the data bears it out we can save so many lives in the upcoming weeks and we are missing that opportunity.
I have already made my choice. I am postponing my second dose. I want my second dose. But I am confident that I can wait. And I can only hope that my second dose, which I have just deferred, will go to someone who it will save their life. It will make a totally different world for that family.
You know some could argue that this could be the end of my career. But I could not sleep with myself at night if I didn’t do this. I just know in my heart of hearts that this is something we must do if we are going to save lives.
giving 2 doses to one person over 65, and nothing to another other person over 65
giving one dose each to two 65+ people
We have evidence that total harm prevention under the second case is higher than the first case, and it also makes perfect logical sense, and there are public health officials confirming this.
On the other hand we have tradition, custom, specific tests.
Say vaccines were tested based on administration in a blue room, but weren't tested in a room painted red. Why are you able to assert without tests that the room color is irrelevant, but are asking for proof that giving two people 85% resistance is better than one person with 95%, when all the evidence supports it? What is leading to wanting to strictly adhere to the exact tested procedure? How do you square your disagreement with the public health officials actively investigating this issue in the linked posts?
People who haven't caught coronavirus yet are still likely to catch it and be at risk. That is, new people are still being infected and are likely to be for a few more months yet.
What calculation are you running to say that changes to policy which might increase first-dose protection are not worth making to protect them?
(I upvoted you, and I'm not sure why you're getting downvotes).
> > Or I can give one dose to both of you. And this is what I know. At the very least, one dose is likely to prevent serious illness, hospitalization and death.
We don't know this from testing. We know it from real world data -- we've given a bunch of people their first dose and then we measured what happened.
Imagine you tell people "we're going to give you two doses of this vaccine. You'll have the first dose today, and then you'll have your second dose in X weeks. We know this is safe and effective from the clinical trials. Do you want to go ahead?"
And then, after you've given them the dose, you say "well, actually, looks like vaccine production is pretty slow, and we really need to vaccinate as many people as possible, so we're going to delay giving you your second dose".
There's a strong argument for doing that, but it's not obviously the right thing to do. Informed consent is an important part of healthcare everywhere. And vaccinations are so important that anything that interferes with trust needs to be really carefully examined.
Just so you don't have the wrong idea, Osterholm's suggestion is that everyone who has already been given the first dose already should get the second dose as planned. Only for new vaccinations should we apply the new plan.
I try to read all perspectives. I appreciate that a lot of people believe vaccination is the only way out of the pandemic, but I think it's a mistake to put 'all the eggs in one basket. I haven't noticed much in the way of efforts to help people become healthier so they can have mild cases of COVID-19, COVID-21, COVID-23, etc.
Furthermore, why are the medical industry wasting vaccine supply on people who are already immune due to surviving their infection? If a person loses antibodies after 6 months from a natural infection, why would they be expected to have antibodies for longer from a vaccine?
I haven't noticed many attempts to rationalize vaccinating people who've already had a case of the SARS-CoV-2. The CDC says "Yes, you should be vaccinated regardless of whether you already had COVID-19. That’s because experts do not yet know how long you are protected from getting sick again after recovering from COVID-19." [0] But that's just "trust us we're experts."
This week I learned my hair stylist had a mild case in January. She diligently wears her mask, but was not protected from getting exposed to her symptomatic co-worker. She was symptomatic herself for 5 days, and recovered without getting too sick. (Her husband was not so lucky, but survived his hospitalization.) She said she's waiting patiently until she too can get vaccinated.
My uncle got himself hospitalized in December 2020 or January 2021. He recovered, but just got vaccinated anyways, even though he certainly still had the antibodies. My brother said our uncle was attesting to his faith in in the vaccine - I think the technical term is "virtue signaling". I think my uncle ought to know better than to use vaccines recreationally.
My brother and his wife both had SARS-CoV-2... He was rather sick for a week or two. She was asymptomatic, but tested positive for antibodies. They both got vaccinated, 7 months after their illness/exposure. My brother was presumably coerced by his workplace, she was probably virtue signaling too.
SARS-CoV-1 burned out after two years. Why would this version be any different?
Is not vaccinating people who've already recovered from SARS-CoV-2 the smarter approach to achieving population immunity?
I think you're being downvoted since you are rhetorically asking questions people have already beat to death. The main two answers being (a) each virus behaves differently, so extrapolation should be done carefully, and not just from your personal experience with surviving past flu's (b) there isn't a limit on how much vaccine we can produce in total, so there isn't really a good reason not to get it. That's perhaps like saying the flu vaccine is only half effective (which is true), so I'd rather get sick twice as often, rather than signal that I trust science.
the mRNA vaccines give the body a blueprint for antibodies, whereas actual exposure has the body creating non-uniform reactions. so that means prior exposed people should "use vaccines recreationally", as in "have a medical professional administer the dosages after an appointment".
the singleshot vaccines are not mRNA based and just help jumpstart coverage. the only goal now is coverage, and these increase the delta week over week of how fast and how far that coverage is. it will allow municipalities to rationalized re-opening all services sooner, by simply having a record of how much of the population has been addressed.
there are many people and organizations and medical professionals making efforts to help people become healthier. many people are trying to find specific links and everything they have right now is much worse than "trust us we're experts" because people don't know. fortunately there is a large audience that would prefer certain messaging to come from anybody but an expert, know anybody like that? so take all the supplements and stay active, because it can't hurt right? all the correlations support it. that's the exact approach regarding the vaccines too, there is no use of any of the approved vaccines worldwide that is worse than contracting covid, even asymptomatic covid. if you are of the predilection that taking vaccines is the same as candy flipping then it should be obvious that will guide your thoughts, the goal of the heads of state is to get enough vaccines available as soon as possible, and that's what they are doing.
> Furthermore, why are the medical industry wasting vaccine supply on people who are already immune due to surviving their infection?
Suppose you're running a vaccine clinic. It's infinitely easier to check prospective patients for their bare demographic facts -- age, residence status, etc -- than it is to first administer an antibody test to see if they qualify for the vaccine.
At best, you'd stretch supplies by 1/(1-seroprevalence), so maybe 30-40%. But in turn, you'd need to take multiple steps to verify patients, and you'd likely lose some along the way who would otherwise qualify (just from the hassle or multiple contacts).
If your "normal times" scientific standards need to change for an emergency, then they were never appropriate to begin with.
But here, I do think the scientific standards are OK. What should be changed in an emergency is not scientific rigor, but the amount of resource that we are willing to spend on the problem with an expectation of waste.
With that mindset, throwing money at the problem to fund trials that address a wide range of questions (e.g., 1 vs 2 doses of Pfizer/Moderna, 2 vs 4 vs 6 weeks for booster timing, etc) would have been the way to go.
> This is evidence of what I call magical thinking–an undue focus on the clinical trial design as having incantatory power.
This isn't complicated. Once we have clinical trial evidence, the protocols tested form the standard of care.
With no standards of care (i.e. no tested vaccine, here), then we can only guess about what might be better than nothing. It's fair to make that decision on a balance of probabilities -- try something if it's 51% likely to help (absent negative side-effects, anyway).
But once we have such a standard, the very deliberate conservatism of our medical science means we need compelling evidence before supplanting it. I think it's very likely that a single dose of Pfizer/Moderna is adequately protective in the medium term (weeks to months), but this was not fully tested. It's not appropriate to start experimenting on the general population based on a hunch.
This community is IT-centric, so think of it like experimenting on a production system. Even if you are 80% sure that something's likely to work out, the risk of failure is so catastrophically high that you don't do it when any other option is open to you.
I think that the most murderous conservative decision was to not allow early human challenge trials on the ethical grounds while hundred thousands of people were dying.
> It's not appropriate to start experimenting on the general population based on a hunch.
By varying the parameters of the situation, could you be induced to change your view? i.e. vary CFR up to 1%, 10%, 50% - given that you acknowledge that the natural experiments we are observing from single-dose administration suggest most of the immunity granted by a vaccine is in the first dose - at what point do you budge?
If you admit that for a 50% CFR, you would not insist on following the established standard of care, then the question is where is the border for you, where your recommendation flips, and you consider early end to quarantine, reduced life loss, to be worth moral hazard of ad-hoc decisions, or not following the standard of care would bring?
Around the same time that you dispense with phase 3 trials entirely. It was more likely than not that the Pfizer/Moderna vaccines were effective by the time they'd completed phase 2 studies. If that is our standard, then we ought to have begun vaccinating at-risk individuals on a volunteer basis then -- in Fall of 2020.
To me, it's inconsistent to both say we need a full phase 3 clinical trial to grant authorization to a vaccine, but then to ignore the protocol of that phase 3 trial when distributing it.
Alternative is to have a more detailed and parameterized protocol.
Regardless, the real solution should have been early human challenge trials and we would have had the vaccine already in the beginning of the summer most likely.
Here's the thing... they did budge. Alex is so caught up in dunking on the FDA that you might miss the fact they just approved a one-shot... within days of hsi last article dunking on them for not doing it.
Presumably, that approach was being studied all along (in the three months that it's even been a decision.
Everyone assumes the FDA are being rigid and bureaucratic regardless of what they actually do. How do you write an article ripping them for magical thinking as they do the thing you're saying they should do? The fact that the FDA didn't pull a u-turn on the same day a"maybe this" thought occurs to a welfare economics professor is enough to accuse them of negligence every week.
Now, it doesn't even seem like even doing what he thinks they should do is enough. They're still caught up in "magical thinking." What for, giving any weight to clinical trial design at all.
Sure, Alex should update his criticism, but it's still valid to look back at why it took this long; First Doses First has been out for a while - it's been 2 months since, for example, Tony Blair publicly advocated for it.
I think the gains we would get from even a single day earlier application of First Doses First probably outweighs the pain caused to the FDA by Alex criticizing them.
It's not about advocating for a position. It's about having a shred of humility, and decency.
What Alex is doing is ragebaiting. That's not noble.
Anyway, it's far from established that either strategy is superior, at this point. There are a ton of factors including execution factors that have nothing to do with clinical trials.
My point is this: It seems like unless the FDA conforms exactly to Alex' opinion on any given day, they are corrupt quacks. It's so arrogant, and a case in point for poisonous discourse. This flavour of rhetoric shots down openness. All you can do is hide behind vague press releases and say as little as possible in public.
If I were working on vaccine development, approval or deployment, I would be trying to stay anonymous. These "open society" people are at the helm of an assault on open discourse.
Alex Tabarrok has been arguing in Marginal Revolution for "First Doses First" since mid-December.
Back then, media commentators and the medical establishment objected "But the clinical trials were for two doses".
Sadly, very few people acknowledged that prudence had a very real cost in additional infections, deaths, risk of mutations, slower economic recovery.
The consensus only shifted to "First Doses First" after the faster spreading UK and South African variants emerged, and the risk of inaction became more obvious.
If we step back from Covid-19, this kind of one-sided thinking is all too common in daily life.
We need more training (at school, university, in media debates) that the choice is never "Do A or not".
It is "do A instead of B (and compare the likely costs and benefits of both)". Or "do the first bit of A, learn some more, and do A' or go back to B". Etc.
My point, expressed more fully, is: "teach people to design projects with measurable outcomes versus the status quo; calculate an NPV to make your implicit assumptions and biases explicit; look for ways to restructure your project plan into intermediate goals where you can reassess your strategy (aka "real options"). And your projects will, on average, cost less and be more successful." My meta-point is: these are useful skills that can and should be taught more widely.
Your argument is essentially "No", is it not?
I'm not sure how that helps improve the situation.
This is basically the primary curriculum of current management training. The more people "train" in it, the more we get to a stupid "ruled by metrics" world.
IRL, you can't actually be scientific outside of certain cases. Bad/fake science can be worse than no science. Fake objectivity is usually worse than subjectivity. The whole things gets bogged down into "searching under the streetlamp" problems.
If you can find a problem that fits the solution you (+ most bogus management consultants of the last 10 years) are advocating, it'll work. Mostly, it gets bolted on to whatever people are doing as an additional layer of bullshit. Nonsense analytics + politics.
Procedural bureaucratic thinking happens to be a very effective way at finding concensus.
Make sure everyone speaks their turn, make sure everyone has a voice. That takes bean counting and nitty gritty arguments. No way around that.
Robert's rules suggests that everyone at a meeting gets two chances to speak on any subject, no more. A big issue with today's discussion is that a few powerful and full of themself individuals take up all the oxygen in the room, removing other people's voices from consideration.
Say your point, then one followup later as other points come up. Then your turn is over and other people talk.
I disagree. Without weighing in on the merits of each strategy too much, I think there is/was a case to be made for each. Also, as you say, consensus did shift. It wasn't a totally fixed mindset. In any case, a decision had to be made. There were always going to be dissidents with a case, regardless of the decision.
I don't think the problem is actually a decision making problem. The 2021 problem is that we're handling disagreements poorly. Whichever decision got made, there would be some dissent.
If you're in a dissident "First Doses First" camp, the other camp is corrupt, pigheaded pseudoscientific bastards that need to go back to school and learn how to think. If/when "the establishment" comes around to your thinking, it's just more proof of this... this article is a case in point.
The establishment side (2 doses, in this case), feels obliged to defend their position, often over defending, making things worse. It quickly devolves into rivalrous rage, both trying to delegitimize the other. Laying piles of bodies at each others' feet. Ad hominems. Accusations of quackery,misinformation and attempts to shut the other up.
You end up here regardless of the decision taken. All roads seem to lead here atm.
>"This is evidence of what I call magical thinking–an undue focus on the clinical trial design as having incantatory power"
Sheesh! Magical thinking... Incantatory power?
What earns the FDA these insults?
In December, they're rolling out two-shot strategy, conforming to the clinical trial design. There is some reason to think that a one-shot first might work better. The "right" answer is complicated, (a) it depends on unknowns that'd need more clinical trials to learn and (b) it depends even more on execution issues related to vaccinated hundreds of millions of people. Different, hard to quantify risks involved in each strategy.
Two-three months later, they're have a one-shot strategy as well. Presumably, this has been in the works for the entire period.
Meanwhile, Alex makes them out to be witches trying to cure covid with willow bark. It's grossly unfair. Maybe they were wrong and you were right, but Magical thinking & Incantatory powers... that is vitriolic bullshit.
I think you and GP have nailed it actually. To use the educational perspective from the top comment, I would want this to be the discussion content for a lesson/section/chapter on the principle of the “existence of candidate decisions” (this is the most dense sentence I’ve ever written).
> Sadly, very few people acknowledged that prudence had a very real cost in additional infections, deaths, risk of mutations, slower economic recovery.
> The consensus only shifted to "First Doses First" after the faster spreading UK and South African variants emerged, and the risk of inaction became more obvious.
But isn’t this exactly what should happen if we’re comparing risks? It seems entirely possible that there was a well-understood risk to giving only 1 shot when the trials were for 2, and that risk was higher than the risk of virus outcomes before the new variants, but then the risk was lower than the risk of virus outcomes after the new variants. I don’t know if this was the case, but it seems very plausible and reasonable. Other things could change these relative risks too, like how quickly we have been able to get vaccinations done.
True. While it is important for the advancement and validation of treatments, Evidence Based Medicine and the quest for "perfect RCTs" has become the new cult for a lot of people
RCTs are a validating procedure, it doesn't mean that people are having zero protection after one dose (of course it is great if we can quantify that and see if it's worth it or not)
As an example, see how for the AZ vaccine the people that got half a dose accidentally at first got (apparently) more protection than from whoever has gotten a first full dose, however, only the full dose was approved (though on the other hand, the J&J vaccine has more virions than the AZ one)
> Sadly, very few people acknowledged that prudence had a very real cost in additional infections, deaths, risk of mutations, slower economic recovery.
Exactly. Now we just need to convince people that maybe, just maybe, it would be a good idea to not have susceptible populations be Vitamin D deficient during a pandemic. You know, just in case.
This criticism of “a quest for perfect RCTs” makes me chuckle. There are so many shit trials out there that sponsors do their best to torture desired results out of its kind of embarrassing.
Instead of saying “prudence has real costs” it would be more accurate to say “having reliable information take time”
The existence of crap trials does not exclude that people are over-critical sometimes, and people take the crap trials that "validate the null (or preferred) hypothesis" at face value frequently as well.
A big risk I don't see flagged is the risk of losing the public's confidence in the vaccine approval process. People are already skeptical of the safety, despite pretty robust trials. Imagine if they went with one dose and it turned out to be marginally effective? Imagine how many people will just opt out of vaccines entirely?
The biggest problem with ALL of these advocates for the "non-trial protocols" is that they want to violate the protocol but nobody wants to put the processes in place to keep an eye out for if something goes wrong.
Read that again: None of the "first doses first" are considering that they might be wrong. And, correspondingly, they have no signals or systems to watch for that.
Okay, this is one thing that ALL the trials did that you want to short circuit. And that's not cool.
For example, what happens if the single dose effects allow a subgroup of people to carry the virus asymptomatically? Oops. You created an immune variant that will evolve to evade all the vaccines as well as lots of spreaders who don't know it.
The human body is stupidly unique and you can get really weird effects. This is why we have trials.
Derek Lowe (https://blogs.sciencemag.org/pipeline/) pointed this out earlier: "first doses first" vs "cache for second doses" has no medical justification--but you're going to have to make a choice anyway. And it's a shitty choice to have to make.
But that's okay. Such is life, you sometimes go with what you have and not what you want--just don't cloak it as a medical decision because you aren't backed by the data. And you have to understand that you might be wrong and you have to have something in place to watch for that.
Many of the earliest voices were calling for single dose trials in December, not just going off trial and doing one dose. I guess you can elide them from your "first doses first" group, but that kind of undermines the whole criticism.
Sure. But that trial takes time and money to run. And you know that the results are going to be inferior. So, the trial is of limited value once you have enough vaccine.
This is the kind of thing that the federal government should have stepped in and funded (one-off, unprofitable, time-limited).
But, in the US, you had a boorish asshat and his seditious bootlickers in charge who promulgated that Covid was a hoax.
You were not going to get something so useful funded by them.
Naive question: given that there is a sort of pareto distribution in covid vulnerability, and that the vaccine rollout prioritizes vaccinating the most vulnerable first, doesn't it also make some sense to prioritize stronger vaccinations (2 doses of Pfizer or Moderna) for the early recipients of the vaccine?
Increasing the speed of the rollout with delayed second doses or skipping second doses sounds compelling for other reasons, but it does seem like a more complex calculus to me than many articles online want to admit.
By delaying the second dose, more of the most vulnerable cohorts can be vaccinated more quickly. By the time you get to vaccinating significantly less vulnerable cohorts with their first dose, you will also be second dosing the most vulnerable cohorts.
So it's not really a case of denying the most vulnerable a second dose in order to give significantly less vulnerable people a first dose.
As an aside, I feel like suggesting this on social media will likely get you banned or in the least, you would have your posts labeled with a warning, and have to endure attacks from others for “not trusting the experts”. First doses first, and I would argue first doses as fast as possible (not having various “equity” focused schemes slowing things down), are very logical stances to hold. In the least, they are reasonable to bring to a societal discussion. The fact that we can’t openly talk about such ideas with friends, coworkers, or broader society is sad.
Way back when, I wasn't convinced about our (UK) vaccination stance; from the start of the rollout it has been first doses - with the gap for the second set at 12 weeks.
The British Medical Association said it was wrong, and there was international condemnation. Now the WHO have endorsed an 8 to 12 week gap. They've also recommended the AZ jab for all age groups over 18, despite French and German comments otherwise (I'm not anti either, I'm a remainer and respect both, but their politician's uninformed comments denigrating the AZ jab were downright dangerous, and the EU is now sitting on unused doses of it).
I don't know whether we've rather unexpectedly got something right in this sorry mess, or just had a very lucky break.
Edit: I'm in the first-doses camp, but not first-doses-first, as the UK is doing two doses in parallel but with an extended gap so a larger volume of first doses go out 'quickly' but people aren't just left with the one. Feels like a good compromise to me.
I also have to say, despite the fact that it is absolutely nothing to do with me, it does feel quite good that the Oxford jab is by far the cheapest and also being manufactured by third parties across the world at pretty much cost. They done well.
65 comments
[ 3.0 ms ] story [ 107 ms ] threadNot an expert, not a researcher, but an economist...
Was that worry all for nothing? Was it backed up by any science? If there was legitimate reason for concern, why do we not have the same concern around the j&j vaccine?
(I'm not trying to advocate any of these points. I'm even less qualified than an economist. But they seem like legitimate questions)
I believe it was because of a lack of evidence either way. The Pfizer trials were two doses. So we didn't know for sure how well one would work. The decision to only do one was made on the basis that this works well for most (but not all) vaccines. It is now known that it indeed works for the Pfizer vaccine, and so it seems to have been a good decision. The concern was that it would not and would set the vaccination programme back by months.
> Why do we not have the same concern around the j&j vaccine?
I'm not sure, but I suspect the J&J vaccine was trialed as one dose from the outset.
The most compelling J&J number is, IMHO, the 100% efficacy at preventing hospitalizations and deaths. Is a single shot of the double-shot vaccines that effective? AFAIK that study was never run..
We have a small sample for the estimate of this efficacy. That being said, all the vaccines have shown this in small samples, so my current working theory is that the hospitalisation rates post vaccine will be very, very low (but almost certainly not 0%).
This is more about looking at the ability of our systems to change their response in emergencies; why are obvious things so hard to do, when they both fully make sense, and are supported by an interested group of experts, against a larger but less interested group of conventional, rule-following experts?
I'm not saying all rejection of convention is correct, obviously - but if you look at the specifics of the argument, not the status or credentials of the arguer, it seems more likely to come to a correct judgement.
Michael Osterholm, Regents Professor, McKnight Presidential Endowed Chair in Public Health, the director of the Center for Infectious Disease Research and Policy (CIDRAP) and state epidemiologist for the Minnesota Department of Health:
https://marginalrevolution.com/marginalrevolution/2021/02/os...In the US there are
- 25 million Americans over 75
- 50 million 65 and older, including the 75 year olds
We have given out 70 million doses.
I think we’re around 40 million jabs a month, hopefully that increases to 60 million soon
There’s no need to experiment.
However, if we find >3-4 weeks between jabs is fine, let’s do it.
On the other hand we have tradition, custom, specific tests.
Say vaccines were tested based on administration in a blue room, but weren't tested in a room painted red. Why are you able to assert without tests that the room color is irrelevant, but are asking for proof that giving two people 85% resistance is better than one person with 95%, when all the evidence supports it? What is leading to wanting to strictly adhere to the exact tested procedure? How do you square your disagreement with the public health officials actively investigating this issue in the linked posts?
People who haven't caught coronavirus yet are still likely to catch it and be at risk. That is, new people are still being infected and are likely to be for a few more months yet.
What calculation are you running to say that changes to policy which might increase first-dose protection are not worth making to protect them?
> > Or I can give one dose to both of you. And this is what I know. At the very least, one dose is likely to prevent serious illness, hospitalization and death.
We don't know this from testing. We know it from real world data -- we've given a bunch of people their first dose and then we measured what happened.
Imagine you tell people "we're going to give you two doses of this vaccine. You'll have the first dose today, and then you'll have your second dose in X weeks. We know this is safe and effective from the clinical trials. Do you want to go ahead?"
And then, after you've given them the dose, you say "well, actually, looks like vaccine production is pretty slow, and we really need to vaccinate as many people as possible, so we're going to delay giving you your second dose".
There's a strong argument for doing that, but it's not obviously the right thing to do. Informed consent is an important part of healthcare everywhere. And vaccinations are so important that anything that interferes with trust needs to be really carefully examined.
This is little better than anti-vaxxing
Furthermore, why are the medical industry wasting vaccine supply on people who are already immune due to surviving their infection? If a person loses antibodies after 6 months from a natural infection, why would they be expected to have antibodies for longer from a vaccine?
I haven't noticed many attempts to rationalize vaccinating people who've already had a case of the SARS-CoV-2. The CDC says "Yes, you should be vaccinated regardless of whether you already had COVID-19. That’s because experts do not yet know how long you are protected from getting sick again after recovering from COVID-19." [0] But that's just "trust us we're experts."
[0] https://www.cdc.gov/coronavirus/2019-ncov/vaccines/facts.htm...
This week I learned my hair stylist had a mild case in January. She diligently wears her mask, but was not protected from getting exposed to her symptomatic co-worker. She was symptomatic herself for 5 days, and recovered without getting too sick. (Her husband was not so lucky, but survived his hospitalization.) She said she's waiting patiently until she too can get vaccinated.
My uncle got himself hospitalized in December 2020 or January 2021. He recovered, but just got vaccinated anyways, even though he certainly still had the antibodies. My brother said our uncle was attesting to his faith in in the vaccine - I think the technical term is "virtue signaling". I think my uncle ought to know better than to use vaccines recreationally.
My brother and his wife both had SARS-CoV-2... He was rather sick for a week or two. She was asymptomatic, but tested positive for antibodies. They both got vaccinated, 7 months after their illness/exposure. My brother was presumably coerced by his workplace, she was probably virtue signaling too.
SARS-CoV-1 burned out after two years. Why would this version be any different?
Is not vaccinating people who've already recovered from SARS-CoV-2 the smarter approach to achieving population immunity?
the singleshot vaccines are not mRNA based and just help jumpstart coverage. the only goal now is coverage, and these increase the delta week over week of how fast and how far that coverage is. it will allow municipalities to rationalized re-opening all services sooner, by simply having a record of how much of the population has been addressed.
there are many people and organizations and medical professionals making efforts to help people become healthier. many people are trying to find specific links and everything they have right now is much worse than "trust us we're experts" because people don't know. fortunately there is a large audience that would prefer certain messaging to come from anybody but an expert, know anybody like that? so take all the supplements and stay active, because it can't hurt right? all the correlations support it. that's the exact approach regarding the vaccines too, there is no use of any of the approved vaccines worldwide that is worse than contracting covid, even asymptomatic covid. if you are of the predilection that taking vaccines is the same as candy flipping then it should be obvious that will guide your thoughts, the goal of the heads of state is to get enough vaccines available as soon as possible, and that's what they are doing.
Suppose you're running a vaccine clinic. It's infinitely easier to check prospective patients for their bare demographic facts -- age, residence status, etc -- than it is to first administer an antibody test to see if they qualify for the vaccine.
At best, you'd stretch supplies by 1/(1-seroprevalence), so maybe 30-40%. But in turn, you'd need to take multiple steps to verify patients, and you'd likely lose some along the way who would otherwise qualify (just from the hassle or multiple contacts).
But here, I do think the scientific standards are OK. What should be changed in an emergency is not scientific rigor, but the amount of resource that we are willing to spend on the problem with an expectation of waste.
With that mindset, throwing money at the problem to fund trials that address a wide range of questions (e.g., 1 vs 2 doses of Pfizer/Moderna, 2 vs 4 vs 6 weeks for booster timing, etc) would have been the way to go.
This isn't complicated. Once we have clinical trial evidence, the protocols tested form the standard of care.
With no standards of care (i.e. no tested vaccine, here), then we can only guess about what might be better than nothing. It's fair to make that decision on a balance of probabilities -- try something if it's 51% likely to help (absent negative side-effects, anyway).
But once we have such a standard, the very deliberate conservatism of our medical science means we need compelling evidence before supplanting it. I think it's very likely that a single dose of Pfizer/Moderna is adequately protective in the medium term (weeks to months), but this was not fully tested. It's not appropriate to start experimenting on the general population based on a hunch.
This community is IT-centric, so think of it like experimenting on a production system. Even if you are 80% sure that something's likely to work out, the risk of failure is so catastrophically high that you don't do it when any other option is open to you.
By varying the parameters of the situation, could you be induced to change your view? i.e. vary CFR up to 1%, 10%, 50% - given that you acknowledge that the natural experiments we are observing from single-dose administration suggest most of the immunity granted by a vaccine is in the first dose - at what point do you budge?
If you admit that for a 50% CFR, you would not insist on following the established standard of care, then the question is where is the border for you, where your recommendation flips, and you consider early end to quarantine, reduced life loss, to be worth moral hazard of ad-hoc decisions, or not following the standard of care would bring?
Around the same time that you dispense with phase 3 trials entirely. It was more likely than not that the Pfizer/Moderna vaccines were effective by the time they'd completed phase 2 studies. If that is our standard, then we ought to have begun vaccinating at-risk individuals on a volunteer basis then -- in Fall of 2020.
To me, it's inconsistent to both say we need a full phase 3 clinical trial to grant authorization to a vaccine, but then to ignore the protocol of that phase 3 trial when distributing it.
Regardless, the real solution should have been early human challenge trials and we would have had the vaccine already in the beginning of the summer most likely.
Presumably, that approach was being studied all along (in the three months that it's even been a decision.
Everyone assumes the FDA are being rigid and bureaucratic regardless of what they actually do. How do you write an article ripping them for magical thinking as they do the thing you're saying they should do? The fact that the FDA didn't pull a u-turn on the same day a"maybe this" thought occurs to a welfare economics professor is enough to accuse them of negligence every week.
Now, it doesn't even seem like even doing what he thinks they should do is enough. They're still caught up in "magical thinking." What for, giving any weight to clinical trial design at all.
I think the gains we would get from even a single day earlier application of First Doses First probably outweighs the pain caused to the FDA by Alex criticizing them.
https://www.businessinsider.com/blair-one-covid-19-vaccine-s...
What Alex is doing is ragebaiting. That's not noble.
Anyway, it's far from established that either strategy is superior, at this point. There are a ton of factors including execution factors that have nothing to do with clinical trials.
My point is this: It seems like unless the FDA conforms exactly to Alex' opinion on any given day, they are corrupt quacks. It's so arrogant, and a case in point for poisonous discourse. This flavour of rhetoric shots down openness. All you can do is hide behind vague press releases and say as little as possible in public.
If I were working on vaccine development, approval or deployment, I would be trying to stay anonymous. These "open society" people are at the helm of an assault on open discourse.
Back then, media commentators and the medical establishment objected "But the clinical trials were for two doses".
Sadly, very few people acknowledged that prudence had a very real cost in additional infections, deaths, risk of mutations, slower economic recovery.
The consensus only shifted to "First Doses First" after the faster spreading UK and South African variants emerged, and the risk of inaction became more obvious.
If we step back from Covid-19, this kind of one-sided thinking is all too common in daily life.
We need more training (at school, university, in media debates) that the choice is never "Do A or not".
It is "do A instead of B (and compare the likely costs and benefits of both)". Or "do the first bit of A, learn some more, and do A' or go back to B". Etc.
Real options improve expected outcomes!
Your argument is essentially "No", is it not?
I'm not sure how that helps improve the situation.
IRL, you can't actually be scientific outside of certain cases. Bad/fake science can be worse than no science. Fake objectivity is usually worse than subjectivity. The whole things gets bogged down into "searching under the streetlamp" problems.
If you can find a problem that fits the solution you (+ most bogus management consultants of the last 10 years) are advocating, it'll work. Mostly, it gets bolted on to whatever people are doing as an additional layer of bullshit. Nonsense analytics + politics.
Make sure everyone speaks their turn, make sure everyone has a voice. That takes bean counting and nitty gritty arguments. No way around that.
Robert's rules suggests that everyone at a meeting gets two chances to speak on any subject, no more. A big issue with today's discussion is that a few powerful and full of themself individuals take up all the oxygen in the room, removing other people's voices from consideration.
Say your point, then one followup later as other points come up. Then your turn is over and other people talk.
I don't think the problem is actually a decision making problem. The 2021 problem is that we're handling disagreements poorly. Whichever decision got made, there would be some dissent.
If you're in a dissident "First Doses First" camp, the other camp is corrupt, pigheaded pseudoscientific bastards that need to go back to school and learn how to think. If/when "the establishment" comes around to your thinking, it's just more proof of this... this article is a case in point.
The establishment side (2 doses, in this case), feels obliged to defend their position, often over defending, making things worse. It quickly devolves into rivalrous rage, both trying to delegitimize the other. Laying piles of bodies at each others' feet. Ad hominems. Accusations of quackery,misinformation and attempts to shut the other up.
You end up here regardless of the decision taken. All roads seem to lead here atm.
>"This is evidence of what I call magical thinking–an undue focus on the clinical trial design as having incantatory power"
Sheesh! Magical thinking... Incantatory power? What earns the FDA these insults?
In December, they're rolling out two-shot strategy, conforming to the clinical trial design. There is some reason to think that a one-shot first might work better. The "right" answer is complicated, (a) it depends on unknowns that'd need more clinical trials to learn and (b) it depends even more on execution issues related to vaccinated hundreds of millions of people. Different, hard to quantify risks involved in each strategy.
Two-three months later, they're have a one-shot strategy as well. Presumably, this has been in the works for the entire period.
Meanwhile, Alex makes them out to be witches trying to cure covid with willow bark. It's grossly unfair. Maybe they were wrong and you were right, but Magical thinking & Incantatory powers... that is vitriolic bullshit.
I think you and GP have nailed it actually. To use the educational perspective from the top comment, I would want this to be the discussion content for a lesson/section/chapter on the principle of the “existence of candidate decisions” (this is the most dense sentence I’ve ever written).
Well case in point https://febs.onlinelibrary.wiley.com/doi/abs/10.1111/febs.15...
> The consensus only shifted to "First Doses First" after the faster spreading UK and South African variants emerged, and the risk of inaction became more obvious.
But isn’t this exactly what should happen if we’re comparing risks? It seems entirely possible that there was a well-understood risk to giving only 1 shot when the trials were for 2, and that risk was higher than the risk of virus outcomes before the new variants, but then the risk was lower than the risk of virus outcomes after the new variants. I don’t know if this was the case, but it seems very plausible and reasonable. Other things could change these relative risks too, like how quickly we have been able to get vaccinations done.
RCTs are a validating procedure, it doesn't mean that people are having zero protection after one dose (of course it is great if we can quantify that and see if it's worth it or not)
As an example, see how for the AZ vaccine the people that got half a dose accidentally at first got (apparently) more protection than from whoever has gotten a first full dose, however, only the full dose was approved (though on the other hand, the J&J vaccine has more virions than the AZ one)
> Sadly, very few people acknowledged that prudence had a very real cost in additional infections, deaths, risk of mutations, slower economic recovery.
Exactly. Now we just need to convince people that maybe, just maybe, it would be a good idea to not have susceptible populations be Vitamin D deficient during a pandemic. You know, just in case.
Instead of saying “prudence has real costs” it would be more accurate to say “having reliable information take time”
And sometimes they don't.
The biggest problem with ALL of these advocates for the "non-trial protocols" is that they want to violate the protocol but nobody wants to put the processes in place to keep an eye out for if something goes wrong.
Read that again: None of the "first doses first" are considering that they might be wrong. And, correspondingly, they have no signals or systems to watch for that.
Okay, this is one thing that ALL the trials did that you want to short circuit. And that's not cool.
For example, what happens if the single dose effects allow a subgroup of people to carry the virus asymptomatically? Oops. You created an immune variant that will evolve to evade all the vaccines as well as lots of spreaders who don't know it.
The human body is stupidly unique and you can get really weird effects. This is why we have trials.
Derek Lowe (https://blogs.sciencemag.org/pipeline/) pointed this out earlier: "first doses first" vs "cache for second doses" has no medical justification--but you're going to have to make a choice anyway. And it's a shitty choice to have to make.
But that's okay. Such is life, you sometimes go with what you have and not what you want--just don't cloak it as a medical decision because you aren't backed by the data. And you have to understand that you might be wrong and you have to have something in place to watch for that.
This is the kind of thing that the federal government should have stepped in and funded (one-off, unprofitable, time-limited).
But, in the US, you had a boorish asshat and his seditious bootlickers in charge who promulgated that Covid was a hoax.
You were not going to get something so useful funded by them.
Increasing the speed of the rollout with delayed second doses or skipping second doses sounds compelling for other reasons, but it does seem like a more complex calculus to me than many articles online want to admit.
So it's not really a case of denying the most vulnerable a second dose in order to give significantly less vulnerable people a first dose.
The British Medical Association said it was wrong, and there was international condemnation. Now the WHO have endorsed an 8 to 12 week gap. They've also recommended the AZ jab for all age groups over 18, despite French and German comments otherwise (I'm not anti either, I'm a remainer and respect both, but their politician's uninformed comments denigrating the AZ jab were downright dangerous, and the EU is now sitting on unused doses of it).
I don't know whether we've rather unexpectedly got something right in this sorry mess, or just had a very lucky break.
Edit: I'm in the first-doses camp, but not first-doses-first, as the UK is doing two doses in parallel but with an extended gap so a larger volume of first doses go out 'quickly' but people aren't just left with the one. Feels like a good compromise to me.
I also have to say, despite the fact that it is absolutely nothing to do with me, it does feel quite good that the Oxford jab is by far the cheapest and also being manufactured by third parties across the world at pretty much cost. They done well.