We should administer Covid-19 vaccines as fast as possible, not reserve doses | Hacker News Reader

We should administer Covid-19 vaccines as fast as possible, not reserve doses

theglobeandmail.com

170 pointselsewhen195 comments

Israelis give shots to medical workers and oldest citizens first. This should drive down the death rate pretty fast. If this approach works, Covid-related deaths in Israel should visibly drop by the end of January.

All countries are doing these groups first, of course Israel is probably the country doing it the fastest so I agree, we should see the results faster.

And I agree with the topic, better 2 people somewhat immune "now" than half with the full vaccine immunity.

> And I agree with the topic, better 2 people somewhat immune "now" than half with the full vaccine immunity.

the 90%+ effectiveness is only with 2 does, who knows how long and if it works with only one dose... it's likely not half and/or not for very long...

Immunity is not binary. Introducing the spike protien to our immune systems now will prime us for a stronger fight later, even if we are not fully immune.

Moderna's phase 3 includes preliminary data on a single dose, and it appears 80-90% effective at/after two weeks. See table 15 on page 28:

https://www.fda.gov/media/144434/download

https://www.wbur.org/commonhealth/2020/12/18/coronavirus-vac...

From the same report:

> The small, non-random sample and short median follow-up time limits the interpretation of these results. There appears to be some protection against COVID-19 disease following one dose; however, these data do not provide sufficient information about longer term protection beyond 28 days after a single dose.

Obviously it'd be great if we can just dose everyone we can once then do a 2nd dose later, but we just don't know enough about how well that would work out long term.

The CDC tried to have non-medical essential workers ahead in line of seniors.

The argument was that, although more people would die (including more African Americans), it would even out black vs white ppl deaths metric.

[edit] source:

https://www.cdc.gov/vaccines/acip/meetings/downloads/slides-...

I don’t see any arguments in those slides about trying to “even out black vs white ppl deaths metric.” Social inequality is one of several factors included in their ethical and logistical plan, but interpreting that as an implied black vs white issue seems like a very uncharitable view of the author’s intent.

Page 31, table row "mitigate health inequities" says:

An upside (green) of vaccinating essential workers (non-healthcare) first: "Racial and ethnic minority groups disproportionately represented in many essential industries. ~1/4 of essential workers live in low-income families."

A downside (red) of vaccinating adults age >65 first: "Racial and ethnic minority groups under-represented among adults >65".

The next page (32) concludes that "mitigating health inequities" (what the paper calls Ethics) is the metric with the biggest difference in outcome, in favor of non-essential workers. A bigger difference in outcome than the one for "maximize benefits and minimize harms" i.e. preventing the largest amount of deaths.

Of course, the presentation doesn't define how these two different things are supposed to be compared. But its recommendation is that "Ethics" outweighs "Science" (their terms, not mine), and therefore populations with larger ethnic minority and low income representation should be vaccinated ahead of those at higher risk from the disease.

This is one of the better critiques I’ve found of the CDC’s very absurd position:

https://www.persuasion.community/p/why-im-losing-trust-in-th...

Literally their plan was to kill a greater number of people “because ethics”. It goes against well established pillars of ethical logic against “leveling” outcomes by intentionally making things worse for a group that is better off, and against the most fundamental tenant of medicine to do no harm.

You must apportion a limited supply of life saving medicine to the population most likely to benefit from it. To claim that apportioning the medicine to a population which is orders of magnitudes less likely to benefit from it, knowing that a greater number of people will die as a result, is an abject ethical failure.

Even accounting for the smaller minority share of the elderly population, you still kill more minorities overall by apportioning the vaccine to essential workers ahead of the elderly, because the death rate is so heavily skewed based on age.

Be warned Yascha has become (or always was, I'm not sure) a grifter hack. He jumped on the cancel culture grift now he's joining in with other armchair epidemiologists such as Matty Yglesias and Nate Silver in misreading slides and propagating a made up controversy about vaccine distribution. The ACIP meeting was recorded and open to journalists. They could go review it, but strangely they didn't. I don't see them reporting video from the meeting where they say anything of the sort your're suggesting.

EDIT: This whole thing is nothing but semi-sophisticated click bait and this is objectively untrue:

> "When the CDC is willing to kill thousands of people in the name of social justice, it's hard to know whom or what to trust.

> So I wrote about its shocking failure of judgment, and my crisis of faith in America’s institutions, for @JoinPersuasion." -- Yascha Mounk, Dec 23, 2020 [1]

Others here have already explained this and shown the priority list from CDC.

[1] https://twitter.com/yascha_mounk/status/1341866668528717824

They don't say ethics outweighs science, do they? They're just two axes on which to evaluate the model. You're coming to some interesting conclusions from those slides, unless I'm misreading something.

I can see where the poster got confused, the powerpoint slides have some sloppy logic. Though whether that’s because of the people or because of the loose thinking caused by slides, or both, is hard to say.

They do not say ethics outweighs science. Dooling simply ignores the science[0] (slide 23) and gives equal "maximize benefits and minimize harms" merit (3 pluses) to vaccinating 65+ and vaccinating essential workers, despite recognizing on slide 21 that with a disease-blocking vaccine, "Initially vaccinating age ≥65 in Phase 1b averts approximately 2–6.5% more deaths, compared to targeting high-risk adults or essential workers".

The tipping point then comes from the equal-weighted (why?) metric of "Mitigate health inequities", where she gives essential workers three pluses but 65+ only one plus. This gives "essential workers" 9 pluses total, greater than the 6 pluses for 65+.

It is frankly shocking that a person can recognize one course of action will clearly save the most lives, then assign an arbitrary score that puts that course of action as equally beneficial as the "woke" course of action--for no reason--and then invent another arbitrary category of equal importance that gives more weight to the "woke" course of action because there aren't enough brown people in the 65+ bucket.

It's even more shocking that this fundamentally irrational and (poorly) racially-motivated reasoning was unanimously approved[1] by the committee.

[0]: https://www.cdc.gov/vaccines/acip/meetings/downloads/slides-... [1]: https://www.cdc.gov/mmwr/volumes/69/wr/pdfs/mm6950e2-H.pdf

You're right, my comment was partly wrong (and it's too late to edit it). They don't say that Ethics outweighs Science as a subject matter. They say that Science predicts all three proposals are equally good, and therefore Ethics gets the deciding vote.

(I disagree with the claim that the choice of which group to vaccinate first won't strongly affect the total number of infections and deaths. But that's a different argument.)

And then, within the Ethics group, they say that the deciding ethical factor in favor of vaccinating essential workers is that they have a higher proportion of minorities and low-income families than the other groups (of high-risk and of old people).

They don't give any reasoning, so it's hard to argue with this. Unlike the Science section, which links a study that models deaths prevented by targeted vaccination. The Ethics section lists some unsupported and unquantified claims (table on slide 31) and at the same time judges which outcomes are better.

Crucially, the first line of this table says that ethically it's equally good to either "Preserve services essential to the COVID-19 response and overall functioning of society", or to "Reduce morbidity and mortality in persons with highest burden of COVID-19 hospitalization and death". Why? Based on what refutable data or model or ethical theory? Who knows.

Of course, this is a summary presentation; there may have been something behind it that they didn't refer or link to.

Adults in the US get universal single-payer health insurance at age 65, and that group is disproportionately white. They also point out that access to healthcare is required for a diagnosis that would place someone in the high-risk group. This is something you'd want to consider if you are in any way interested in evaluating institutional racism.

I think it is entirely fair to consider whether a policy institutionalizes that a person at high risk because of age is more important to save than person at high risk because of undiagnosed condition. .... or whether the groups that society expects to go to work in person should be protected over groups who have already been given social benefits that are intended to enable them to avoid going to work.

This is, as the powerpoint outlines, a purely ethical, and non-scientific evaluation of the situation. The scientific arguments are in their respective sections.

> But its recommendation is that "Ethics" outweighs "Science" (their terms, not mine)

I don't see this anywhere in the document.

"“Older populations are whiter, ” Harald Schmidt told the New York Times. “Society is structured in a way that enables them to live longer. Instead of giving additional health benefits to those who already had more of them, we can start to level the playing field a bit.”

See https://www.nytimes.com/2020/12/05/health/covid-vaccine-firs...

I thought that the point of vaccinating essential workers first is that they are those that are more likely to spread the virus so by starting with them you stop the virus before it reaches the seniors.

Depending on the factors (% of seniors vs. essential workers, availability of vaccine) I can see either approach being more effective.

Not aware of evidence that vaccination stops or reduces spread. Given that asymptomatic and presymptomatic people are the main spreaders, it's certainly not a given.

Vaccinating front line health and care workers does make sense for two reasons

1) Far more likely to catch it

2) Far more problematic if they catch it and are off ill because of it

Other front line workers like police probably less important -- less likely to catch it, and not as problematic if large numbers are off ill as the people who fix it. Saving lives and reducing load on hospitals by vaccinating the over 70s is likely more beneficial.

There's a 3rd reason. Healthcare workers are frequently working directly with high risk people.

An asymptomatic infection of a healthcare worker could lead to a typhoid mary situation. Especially if they are in the right area of interaction (geriatric care, oncology).

Firstly healthcare workers should be being tested on a very frequent basis to catch for asymptomatic cases (as well as tons of PPE), secondly there's no evidence the vaccine will stop that type of transmission.

> Firstly healthcare workers should be being tested on a very frequent basis to catch for asymptomatic cases.

Agreed. It's a good idea even if they have been vaccinated until community spread is way down.

> secondly there's no evidence the vaccine will stop that type of transmission.

Why do you say that? The vaccine should trigger a strong immune response which should keep most people from getting infected. That's my assumption anyways.

Seems like the multi-prong approach here is the most helpful anyways. Vaccine + testing + masks to eliminate spread as much as possible in vulnerable communities.

> Not aware of evidence that vaccination stops or reduces spread.

There isn't any evidence for the opposite, either. There is no evidence either way. You can't just choose between two unknown options based on which one you like, and then defend your position by saying that there is no evidence against it.

There is no evidence the sun will rise tomorrow either.

But it is still rational to use logic and experience and assume that sunrise will probably happen as usual tomorrow, and these vaccines probably behave like other vaccines.

You think there is no evidence the sun will rise tomorrow?

I'm really trying to make the point that there is evidence that vaccination reduces spread, if you open your mind a little.

My use of sarcasm is maybe not the best way to convey that, but I am who I am.

I think the near eradication of multiple devastating diseases like polio, smallpox, and the measles is evidence that vaccination works and reduces spread.

Seems that HN Doesn't like the idea that there is no evidence about how the vaccine works with spread

https://www.huffingtonpost.co.uk/entry/getting-the-vaccine-d...

Professor Chris Whitty, England’s chief medical officer, recently told a press conference that true herd immunity will only occur if we have vaccines that can reduce transmission between people. At the moment, we don’t know if they do.

https://www.abc.net.au/radio/programs/pm/who-warns-vaccines-...

But the World Health Organisation says even though vaccines have been created.. none of those will necessarily stop the spread of the virus, rather - only stop the virus from becoming a serious disease inside our bodies.

https://www.independent.co.uk/news/world/americas/coronaviru...

"They do not show that they prevent you from potentially carrying this virus transiently and infecting others," Moderna Chief Medical Officer Tal Zaks told Axios, adding the public should not “over-interpret the results” of the vaccine yet.

I'd defer to Moderna's CMO, Englands CMO, and the WHO, but it does make sense -- after all the vaccine helps you fight the virus, but it seems people are spreaders before they even have symptoms (which I believe is a sign of the body engaging it)

Aiming a vaccination program at spreaders rather than those most likely to suffer seems a risky move.

Here's a more recent presentation from the ACIP working group: https://www.cdc.gov/vaccines/acip/meetings/downloads/slides-...

The order the working group currently proposes:

1a: health care personnel, long-term care facility residents

1b: frontline essential workers, persons aged 75 years and older

1c: persons aged 65-74 years, persons aged 16-64 years with high-risk conditions, essential workers not recommended in phase 1b

2: all people aged 16 years and older not in phase 1, who are recommended for vaccination

The parent comment is also mischaracterizing the reasons the presentation gives for putting essential workers at the same priority as persons aged 65-74. Here's what it says (page 33):

* Essential Workers are at high risk of exposure. Prevention of disease will reduce transmission

* Preserves services essential to the COVID-19 response and overall functioning of society. “Multiplier effect”

* Workers unable to work from home

* High level of interaction with public or others in the workplace

* May be unable to control social distancing

* Frequently interact with others in the workplace

* Racial and ethnic minority groups disproportionately represented in many essential industries

* ~1/4 of essential workers live in low-income families

There’s actually a lot of variety in the priority order. Other than healthcare workers the following groups will be prioritized above age 65+ in Massachusetts

1. First responders

2. Nursing home residents

3. Homeless

4. Incarcerated

5. 2+ comorbidities

6. Top priority essential workers

7. Teachers

8. One comorbidity

(Check out your state here https://www.washingtonpost.com/graphics/2020/health/covid-va...)

40% of all deaths in the US have been nursing home residents.

I've seen this number around for a long time, but since I never saw a source posted, I don't know if this is still true.

Edit: I found https://covidtracking.com/analysis-updates/ltc-deaths-pass-1... for December 2, which shows about 100,000 deaths from "long-term care facilities". Not all states report differences between staff and residents dying, but I think it's reasonable to estimate it's mostly residents. On that day they show a total of 265,000 deaths in the USA, so about 37% from LTC.

My mother got her first vaccination shot yesterday. She's in a nursing home in Virginia. She's already had one COVID exposure (luckily she didn't get it), and it's just scary there.

they're also administering these shots out at a much faster rate relative to delivery whereas we've only administered less than 20% of the doses that have been delivered so far (11.4 million delivered, 2.1 million administered). The delivery rate should only increase especially even as there are several more vaccines in the pipeline so being able to actually administer the delivered doses may be the bottleneck moving forward.

Aren't we doing the same here in the US?

Mostly, each state decides what to do with vaccines though, so it's not as uniform.

I just spoke with some people in Israel. None of whom are healthcare workers. They all got the first shot of the vaccine. ~4 people.

Were they older citizens?

Israel has opened up vaccinations for healthcare workers and anyone above the age of 60.

There are a few other groups getting vaccinations as well - I don't know all of them, but I know of some people working in nursing homes or other similar jobs.

In addition, there are people hanging around the vaccine administration centers at the end of their workday, and hoping that they have leftover vaccinations that would otherwise expire. Apparently, some people who otherwise wouldn't get vaccinated are able to get the vaccine that way.

Sort of... there are still plenty of healthy people that only didn't die due to intubation.

I'm curious to see the numbers there as well. If we vaccinate the old, then resume life as normal, we're going to overwhelm intubation machines.

I'd read medical studies and news articles over the summer that intubation was causing worse outcomes than things like proning in many cases, with dramatically improved survival rates in hospitals that avoided intubation. I'm wondering if advice on that has changed at this point. There was also an interesting piece on HN about jailbreaking CPAP machines to increase supply, although that might not be necessary now?

The standard of care has evolved based on the data. Unlike other respiratory infections, intubation often provides no benefits and some harm to those COVID patients who are not literally dying of asphyxiation (critically low O2 saturation). Basically intubation is now a last-ditch effort, that can and does save lives. But you'd better hope you don't need it because the likelihood of needing intubation AND surviving it are low.

Israel's deployment is amazing, they went from 0 to 8% of their population in 10 days. How many doses have they actually physically got?

Summary:

The title makes it seem like they are talking about medical professionals vs general public.

But actually they are talking about reserving second doses for those who got the first dose, and the article says it's better to actually administer those doses, even if it means a delay on the second dose for people.

I hope that if people don’t bother to read the article they at least see your comment. Most of the comments here so far are debating the merits of what they assume from the title rather than the text of the article.

When a pay-walled story is posted, the number of people who read the story drops significantly.

Thanks for this quick summary. I've got a pay-wall error and can't read the article.

So what happens if the second dose arrives very (eg. many months) late? Can you start again and give 3?

And is there any chance all the partial immunities will catalyze resistant mutations?

Disclaimer: not an immunologist.

> So what happens if the second dose arrives very (eg. many months) late? Can you start again and give 3?

WHO guidance on incomplete vaccine courses (pre-COVID): https://www.who.int/immunization/policy/Immunization_routine...

In almost all cases, the suggestion is "resume without repeating the previous dose". The only exceptions are typhoid and cholera vaccinations. I don't know what these suggestions are based upon, but the source is as close to expert consensus as it can go (the WHO has gotten a lot of flak on COVID, but I have never heard anyone seriously doubt its guidance on standard vaccinations).

> And is there any chance all the partial immunities will catalyze resistant mutations?

Apparently this doesn't even happen much for antibiotics ( https://jamanetwork.com/journals/jamainternalmedicine/articl... ), so I wouldn't worry about it happening for vaccines. See https://www.pnas.org/content/115/51/12878 for a comparison. If it were a major issue, it would just as well be an issue with fully delivered vaccination courses, as they don't offer 100% immunity either. The amount of mutations arising should still be proportional to the sick population (or, rather, to its total prevalence over time and population), and anything that reduces the latter should reduce the former. Any vaccine-immune mutation can just as easily come up before the vaccine, unless it hampers the virus so strongly that it is unviable until the evolutionary landscape has shifted. At that point, however, we're looking at a much simpler problem.

> So what happens if the second dose arrives very (eg. many months) late? Can you start again and give 3?

Stockpiling of vaccines is a method to obtain certainty that you can vaccinate on schedule. There are better methods of ensuring certainty, e.g. by supporting manufacturers as well as possible to meet schedules, analyze weaknesses in supply chains, trying to counter them early, etc.

The issue is the cost of maintaining the stockpile:

> she and her colleagues projected that frontloading vaccine doses would avert between 34 and 42 per cent more symptomatic coronavirus infections, compared with a strategy of keeping half the shipments in reserve.

What's the cost of delaying the second shot for some groups? It's certainly non-zero, but 34 to 42 percent less symptomatic infections is a lot of prevented infections (and death).

The cost-benefit analysis gives a clear conclusion: don't stockpile.

you described precisely the reason why anyone wielding deciding power on the matter and acting rationally wouldn't allow the second doses be administered until [almost] everybody (in the country or on the whole planet) gets the first one. To prevent such a situation the subjects of that power have to stockpile.

The last year has been supply chain disruption after supply chain disruption.

If the vial manufacturer has a covid outbreak at their facility and has to stop manufacturing for two weeks, your scheduled shipments aren't going to be on time.

As supply goes up, and we get past the priority recpients to the general public, it probably makes sense to reduce the stockpile requirement; some general public aren't going to come back for their second dose anyway, and more risk is appropriate for general public than high risk groups.

> The last year has been supply chain disruption after supply chain disruption.

Indeed, but even if 20% of people don't get their second shot on time and the impact of that is equal to them not being vaccinated at all (we don't know but likely the impact is smaller), even then the impact would be lower than stockpiling vaccines which increases the number of symptomatic infections by 34 to 42 percent.

> If the vial manufacturer has a covid outbreak at their facility and has to stop manufacturing for two weeks, your scheduled shipments aren't going to be on time.

That's precisely where the government can step in and ensure that antibiotics therapies are available to workers in those facilities as well as daily tests. Then it doesn't have to shut down.

> As supply goes up, and we get past the priority recpients to the general public, it probably makes sense to reduce the stockpile requirement

At that point it's less relevant which strategy you choose.

> Indeed, but even if 20% of people don't get their second shot on time

I think my worst case expectation is much worse than yours.

With a two week production shut down, I expect my corner of the world would not get new shipments for more than two weeks; higher priority places would be getting their previous orders filled before us, and we'd only get more once the backlog was filled. Also, I fully expect any production will take at least a week to be made public, so too late for an eager dosing regimine to start holding back.

Keep in mind, we're in winter, and winter storms often cause logistical problems.

The one dose efficacy data is too low confidence to make good decisions with. I'm open to the argument that it's better to use two doses to give two people one dose than to give one person two doses, but it would have been nice if that was trialed. There's also a related question of if you have a dose, if it's better to give that to a properly timed first dose recipient, or to give it to someone who hasn't received a dose yet. Again, there's not quality data; I would lean towards following the trial parameters.

> That's precisely where the government can step in and ensure that antibiotics therapies are available to workers in those facilities as well as daily tests.

Where has this suddenly competent government been for the past year? Yes, we expect a new executive in the US in a month, but the handoff takes time in normal conditions, and we have reason to believe this one will be more abrupt than many.

How do antibiotics help for covid?

How do daily tests help avert a shutdown when you can spread the disease while testing negative? One person could infect most of a shift of workers, given the right circumstances; more than one shift if they do handoff work to the next shift.

Supply chain workers for the vaccine aren't in the priority lists for the vaccine either. Anyway, covid isn't the only communicable disease possibility.

> higher priority places would be getting their previous orders filled before us

Hmm yeah that's a problem with the vaccine manufacturer. Ideally in such a scenario, the second doses get served first, after which the higher priority first doses get served.

> How do antibiotics help for covid?

Urgh my fault sorry, I meant antibodies not antibiotics. I tried to use them as example for something that has limited availability but allows to shorten disease durations.

> How do daily tests help avert a shutdown when you can spread the disease while testing negative?

Of course a shutdown is most effective and if done on a population scale it's bad for the pandemic. But daily tests do help to curb the spread.

I would bloody well hope that people manufacturing COVID vaccines are the quintessential essential workers, who should be the first in line for getting COVID vaccines.

On the other hand, given how poorly essential workers are actually treated by our society, I doubt that's how things actually work. (See - senior hospital administrators getting vaccines ahead of front-line medical staff...)

And their supply chain, but now we are at everybody anyway.

Sometimes taking chances is worth it. As far as I have heard, every single vaccine trial has concluded “yeah, our vaccine is effective.” So if people had been allowed to take experimental vaccines, the pandemic would have been a lot better.

In this circumstance as well, taking a chance is worth it. Even if the virus mutates, changing the mRNA vaccines shouldn’t take more than a week.

> So if people had been allowed to take experimental vaccines, the pandemic would have been a lot better.

> In this circumstance as well, taking a chance is worth it.

Nope nope nope nope. This is the literal example of survivor bias. You can't look at the results and use them to inform prior behaviour. The vaccine could have had catastrophic side effects.

I think it is clear the regulatory framework wasn't ideal.

The last stage of the trial (which took three months) was not about if they were safe - but simply if they worked. Meaning, they were cleared to have minimal side effects, just unclear if it worked.

If you had approved people to choose to take the vaccine in August we could have had 50M people with shots already.

A lot of people have my skepticism that the main reason that the FDA didn't want to do that since it might be a bit of a media heartache for them, but so what. Even if you thought it would only work 20% of the time that 20% of the time it saves 100,000+ lives.

50M doses isn't something that gets suddenly summoned. It's a big logistics problem. Your assumption is valid only if the pharmas have capacity to make a extremely big amount of vaccine doses. That costs money and (I guess) materials, and both of those things could go towards a vaccine that is proven to be effective.

They had the vaccines in storage in October. Not as many as they had in December but they were just awaiting approval. The rollout would have been slower then with less to go around - but a headstart on the vulnerable population or potential spreaders would have been significant.

I think you misunderstand the clinical trial approval process. Yes, phase 2 is primarily about safety, but these are much smaller trials than phase 3, which in turn is much smaller than the post-marketing patient population (cohort). Safety issues routinely arise in phase 3, or even after, due to the much larger cohort sizes. Good phase 2 results are not proof of safety, rather an indication of safety. It is bordering on genocidal levels of reckless endangerment to release a vaccine for general use after Phase 2 in the hopes that it might be effective and is probably safe.

Whether or not it would have been a full three months I stand by my belief the current system was not designed for the cost-benefit of a raging pandemic.

The three month time of phase III had nothing to do with safety. That long period was only used because it took that long time for enough people to catch Covid naturally for them to prove statistical effectiveness. In the current system, if Covid had been raging at an extreme rate they would have had enough positives in the control group after a month and ended the trial then. If it had been under control, it actually would have taken many more months to declare effectiveness even though the extra wait time would have shown no more data on safety.

Whether or not phase II was sufficient on its own they could have declared the vaccine probably safe a month after phase III began instead of after three months. This would have still saved 50,000 lives in the US at minimum.

All first 5 to be developed vaccines are pretty successful. So don't think survivor bias applies here. Also, what exactly are those catastrophic side effects? All the cases of "vaccine gone wrong" are pretty mild and most of them don't apply to BionTech/Moderna vaccines (since it doesn't include virus) or related to product problems, which don't depend on trials. If vaccines were started to be used after phase 2, thousands and thousand of lives could be saved. I think it's worth discussing. Especially with these new mRNA vaccines, where new one can be created within a month, it could be a new way to battle upcoming pandemics.

They had to pull a few because of side effects during the trials and some just not working. Producing vast amounts before they have been tested at all would have been costly when they had to throw them out.

Due by the first 100 000 recipients, 3 people had an allergic reaction. Those 3 people still received a second dose.

It would not only be costly to throw out the vaccines, but also stupid.

Do you have a full list of reactions? Where can we find that informatiin?

Sorry, I don't recall the source for this information as it has been a few weeks since I came across it.

> Producing vast amounts before they have been tested at all would have been costly when they had to throw them out.

I'm under the impression that's precisely what the world governments did? There were bar charts going around saying that XX% of each country's supply commitments came from Pfizer or Astrazeneca or Moderna or whatever.

>Producing vast amounts before they have been tested at all would have been costly when they had to throw them out.

This is exactly what they did - funded by governments.

Example:

https://en.wikipedia.org/wiki/Operation_Warp_Speed

>The program promotes mass production of multiple vaccines, and different types of vaccine technologies, based on preliminary evidence, allowing for faster distribution if clinical trials confirm one of the vaccines is safe and effective. The plan anticipates that some of these vaccines will not prove safe or effective, making the program more costly than typical vaccine development, but potentially leading to the availability of a viable vaccine several months earlier than typical timelines.

"allowed"

Do you think there were a bunch of experimental vaccines just floating around on the . . . dark web? ... That's not how it works. The vaccine makers themselves have no interest in distributing unproven vaccines. I myself was highly confident (and posted such on HN) that we would have a successful coronavirus vaccine, based on the scientific information available early in the pandemic. But it wasn't a surety. You must have a randomized trial or you risk very very bad outcomes. Some experimental vaccines (e.g., dengue) actually make the viral infection worse. Some don't work at all (e.g., Sanofi's covid vaccine). Heck Sanofi's vaccine is likely the easiest to manufacture and distribute. If we followed your "plan", it would likely be that most got the ineffective Sanofi vaccine and we would be moving dead bodies around with bull dozers as "vaccinated" people failed to practice social distancing.

I got the Pfizer vaccine in Alberta, Canada. Our vaccine FAQ says that getting the second injection 21d to 28d after the first is recommended. Up to 3 months after is allowed, but not advised. For delays longer than 3 months, the person simply won't get the second dose.

It seems reasonable for different provinces to take different approaches, since their disease patterns vary significantly.

Ontario and Québec have endured many weeks of increasing numbers of diagnosed infections per day, whilst the other provinces have seen decreases.

The absolute incidences also vary between provinces, letting some test-and-trace cases more easily than others.

> Ontario and Québec have endured many weeks of increasing numbers of diagnosed infections per day

Which would indicate that they -shouldn't- be reserving doses...

>..whilst the other provinces have seen decreases.

Nova Scotia had a big jump mostly Halifax since it's the most populated part of the province.

That's interesting phrasing. Ontario & Québec have increasing numbers but they still have lower per capita case rates than western provinces.

Not really true anymore, according to https://mackuba.eu/corona/#compare.daily?pop=1&c=ca-on,ca-qc...

Alberta's numbers over the holiday are nonsense. Hinshaw even said so herself.

I would think we have the models to try many different strategies to determine the "best" distribution methodology. Depending upon what the definition of "best" is, of course.

I would take Alberta's latest numbers with a giant grain of salt. Testing rates are awful there right now.

Too bad my family there used the numbers as an excuse to break quarantine. Enraging.

My sister lives there and knows multiple people her age (mid 20s) who've had the coronavirus. To my knowledge I only know a single person here in the US who has had it. Seems bad.

Alberta is about the middle of the pack in North America for per-capita rates. Not as bad as the worst US states, but nowhere close to the good ones.

Ontario, meanwhile is better than any US state other than Hawaii, last time I looked.

However, if you look at per-capita death rates, Ontario does not look so good.

The system seems kind of weird. My friend works at a hospital in one of those provinces and they are doing a quasi-lottery system to distribute the vaccine to their employees. Users have to login to a system, pick a time beginning at a certain time of day, and if there are no availability well tough luck. My friend was aghast that they weren't prioritizing employees most likely to intubate patients

Seems like there could be a middle ground, administer more or less of the second dose based on the amount stockpiled and the probability of future deliveries.

Testing is already strained? And missing doses is all waste. Too risky imho?

I have heard a single dose is still pretty effective.

Copenhagen hospitals vaccinating more people than scheduled [0]

> Hospitals in the Copenhagen area are ahead of their COVID-19 vaccination schedule, according to the Capital Region’s vaccine taskforce.

> According to the taskforce, 20 percent more people than calculated have been vaccinated because each vial contains at extra 1-2 doses.

> The reason for the extra doses is producer Pfizer/BioNTech adding a bit to account for potential spillage.

> And so to attain the excess doses, health personnel must handle the vaccine with care.

[0] https://cphpost.dk/?p=121536

Edit: formatting

Ontario changes COVID-19 vaccination plans after slowing rollout [1]:

> Ontario has changed its COVID-19 vaccination plan to give a first dose to as many people as possible and no longer hold second doses in reserve

General Rick Hillier (retired) to Provide an Update at Queen's Park [2] today at 11am EST which is currently shown in the upcoming "Live" tab of CPAC [3].

[1] https://www.theglobeandmail.com/canada/article-ontario-chang...

[2] https://news.ontario.ca/en/advisory/59836/general-rick-hilli...

[3] https://www.cpac.ca/en/

The update by Rick Hillier is complete [1]. For Ontario residents, the key new information:

   - clinic holiday shutdown was a mistake; daily operation will now be the norm
   - first Moderna shipment arrives within 24 hrs, will be distributed outside clinics
   - the vaccination rate is currently limited by supply
   - 80K Pfizer / 50K Moderna per week in January
   - Phase 1 (vulnerable and health care workers): Jan-Mar (~1M)
   - Phase 2 (expanded distribution): Apr-Jul (~8M vaccinated total)
   - sentiment: over 80% of population willing to vaccinate as soon as possible
   - Hillier requested that Health Canada consider a 1-dose Moderna regiment

[1] https://youtu.be/dy0ctRiU2IM?t=1008

Somewhat off topic but in theory, would it be possible to have jab #1 be Moderna's and jab #2 be Pfizer? Would it be ethical to test this? Is there anything inherent in either's approach that would make this dangerous?

There only way to know for sure would be to run a controlled trial. Vaccines can work (or not work) in surprising ways.

The danger is that it's untested and might not provide any more benefit than just the first injection.

In theory, yes, that should work. In practice, immunology is very complex and the only way to be sure is to test.

As just another software guy who's near that dangerous "knows enough to be overconfident" spot:

That would probably not work the same way.

The point of a booster shot is to make the immune system "think" it's been infected by the same intruder twice, so it steps up permanent defences to a more serious level.

Both the Moderna and Pfizer vaccine produces replicas of the virus "spike protein", but with (presumably) different mRNA formulas.

So if their different versions are similar enough that the immune system sees them as "this shit again!", which would work as a booster shot, or as an unrelated infection, is impossible to say. At least for me, who is (again) unqualified to talk about this stuff.

That said, two vaccines would at minimum be better than one.

Well, they are at least similar enough to the real thing that the body would say "this shit again!" for the real virus. My guess is that they have different start/stop sequences (that shouldn't matter for the protein the mRNA is encoding) and different stabilizers to help the mRNA survive.

That approach is called heterologous prime-boost, and it's likely being tested for the Oxford/AstraZeneca + Pfizer combo and the Oxford/AstraZeneca + Sputnik V combo to try to increase potency (not to increase supply). It's not being tested for the combo of the two mRNA vaccines (Pfizer/BioNTech and Moderna) though.

That's something that may well get trialled at some point. In the absence of a trial, it certainly wouldn't be ideal, though.

This is an early stage situation. Once the supply pipeline starts running smoothly, having some number of doses in reserve will still be needed, but not as many.

Combat logistics for ammo has the same problem. The person in charge of vaccine distribution planning is an Army logistics general, who will see that.

It's a different problem here-- making sure the second dose is available at all costs.

The problem here isn't a logistics problem, but a "which alternative is worse" problem.

It sure looks like 1 dose is still >50% effective after day 10. The data is sparse: you wouldn't want to give people a single dose on purpose. But holding back more than half the doses in case there's a low probability event of a complete supply disruption seems overcautious. Hold back 25%.

I know second hand that UCSD is giving all doses right now. Vials contain 20% more to account for spills, so they are using this extra to give extra vaccinations. The second dose for each patient will come from future vaccine shipments. Other providers in San Diego may be doing the same thing.

Even if it were the case that the vaccination is only effective once the second dose is given in time, holding back half the supply is far too drastic: if the supply chain failed to the point where these doses would all be used for this purpose, we would have a much bigger problem on our hands.

Berlin today closed it's sole vaccination center. Allegedly because of a lack of interest by the target audience. So we now reserve all doses, not just the second one. Total organizational fubar.

People have been forced to endure lockdown and masks, but they have a choice about vaccines. We should just vaccinate the elderly and anyone else who wants it and get back to 100% normal life - ditch the mask rules and social distancing.

That's hard to believe in multi million city there was not enough people to fill on one center. Maybe there was something other preventing people from going there? Do you have source?

I found it hard to believe as well, but found [1]. Not living there at this time, I lack a sense of current sentiment, but it's my understanding that so far only medical personal is being vaccinated. Allegedly there was a lack of interest of being vaccinated right now (it's Christmas season and many are on holiday there I think). That'll change in a few days.

[1] https://www.tagesspiegel.de/berlin/zu-wenig-rueckmeldung-zu-...

Multiple sources in German/Berlin Media. Take for instance this piece:

https://m.tagesspiegel.de/berlin/zu-wenig-rueckmeldung-zu-we...

As far as I was able to connect the pieces the organization was a total disaster. Mobile teams going into nursing homes do not vaccinate the people working there. These people should have come to the center, but could not because the were working at that time. Other elderly people (not living in nursing homes) should have come to the center but did not receive the invitation in time.

This kind of utter incompetence is, sadly, quite typical in Berlin.

I wish there were a "flag" button but for the ensuing low-effort, low information discussion you know will come for a story clearly outside the expertise of anyone who posts here.

flags as such

If I come back and read any comment here that leaves me with more knowledge than when I arrived, I promise in the future to reward HN by donating 4 horror stories about uncles working union jobs.

There is. Go the direct link for a comment.

I don't get why nobody seriously considers a one-dose regimen for the fastest possible vaccine rollout. All available data for both, Biontech/Pfizer and Moderna's vaccines suggests that one dose is enough for >90% efficacy which would be sufficient to reach herd immunity. There was a very good opinion article in the NY Times [1] about that, but not much more.

The problem here is that a one-dose regimen was not studied in the trials. While it is good medical practice do not deviate from trial designs (see the Oxford/AZ vaccine), the data is very clear here. The vaccines are effective 14 days after the first dose and before the second dose. Sure, there are risks (maybe the efficacy drops later on or it is more difficult to reach the full efficacy if the booster shot is delayed) but these risks seem to be negligible compared to the hundreds of thousands additional unnecessary deaths due to limited vaccine supply and delayed vaccination with a two-shot regimen. We know from other vaccines that the timing of the booster shot normally is not that important and it is very probably that a booster shot after a few months would be exactly as good as a booster shot after 4 weeks.

In my opinion, nobody seriously contemplates the one-dose regimen because somebody has to take responsibility to deviate from the proven and tested protocol - politicians have nothing to win here but much to loose in case anything goes wrong. And big pharma has no interest in financing and starting one-dose trials asap b/c it would hurt their own bottom line and there is no financial incentive at all, quite the opposite (+they have the same responsibility problem). This is a deadlock which will cost many, many lifes.

[1] https://www.nytimes.com/2020/12/18/opinion/coronavirus-vacci...

That's not what the Pfizer paper showed. There was a confidence interval from between 20-70% iirc for the first shot. Maybe if we're lucky it's on the higher end, but I would stick to proven science if we're going to combat this thing effectively.

Wargame the scenarios. You wouldn't want to give one dose on purpose, but reserving a whole bunch of doses to prevent some people from getting only 1 in a worst case scenario is silly.

That 20-70% efficacy number includes the entire period between dose 1 and 2. It looks way better 10 days after dose 1, btw-- MLE around 80%.

e.g. single dose -- 50% efficacy (to be pessimistic); double dose -- 95% efficacy. 10M doses "in flight" monthly. 100 deaths/million/month. Ignore all effects on transmission (to be more pessimistic).

Scenario 1 (no logistics disruption, half held back): 5M dosed with 1 dose at day 0, 5M dosed with 2 doses at 1 month and 5M dosed with 1 dose at 1 month. 250 saved in first month, 725 saved in second month.

Scenario 2 (no logistics disruption, 25% held back): 7.5M dosed with 1 dose at day 0, 7.5M dosed with 2 doses at 1 month and 2.5M dosed with 1 dose at 1 month. 375 saved in first month, 837 saved in second month.

Scenario 3: (100% production disruption, half held back): 5M dosed with 1 dose at day 0; 5M dosed with 2 doses at 1 month. 250 saved in first month, 475 saved in second month.

Scenario 4: (100% production disruption, 25% held back): 7.5M dosed with 1 dose at day 0; 2.5M dosed with 2 doses at 1 month; (5M late for dose 2); 375 saved in first month, 612 saved in second month.

With a 25% reserve, you only get into the unexpected-only-1-dose regimen with a supply disruption of more than 50%... and even then it's still probably better than a 50% reserve.

This is a productive line of thinking.

I wonder if hitting the delivery targets of 20M per month is going to be achieved. That's the first bottleneck before the vaccine supply.

Right now, many jurisdictions are supply limited, and most look like they're ramping enough to be supply limited, though still a bit more slowly than I'd like.

Lowering reserves would still increase administration rate immediately (though not proportionally). Actually, we might even enter phase 1B in some areas and have a mostly parallel administration system start to come up...

It's unclear to me how much of the slowness so far in administration (13.5% of doses allotted are reported as administered so far)... is A) (fixed?) delays in reporting, B) (fixed?) delays in logistics getting doses to administration sites, C) limitations on rate of administration, D) holiday-related slowness. Only C really matters in the long term-- if it is ramping poorly compared to production than that's a problem. But I think it's premature to assume C is limiting when A, B, & D are surely large right now.

Did you see that graph with amount of cases in control/vaccine groups? Where around day 10 vaccine group graph shows massive drop. It looks pretty convincing. And by now they have much more data, especially with lots of cases nowadays. That simple move can save thousands of live and I think worth considering, instead of just ignoring everything.

I just reviewed it and somehow I missed that detail that the inflection point was way before the second vaccine on the first reading. Very interesting.

Note reviewing the paper: 95% CI 29.5% to 68.4%, mean vaccine efficacy prediction 52% for 1st shot.

That is not correct (I think that number comes from the Oxford/AZ vaccine?).

If you look at the cumulative incidence curves (for Moderna see [1] page 28), you can see that COVID-19 occurences drop approx. 14 days after the first dose with no measurable effect of the second dose after 28 days. This pattern is exactly the same for both vaccines.

[1] https://www.fda.gov/media/144434/download

PS: My parent comment was downvoted more often than any other HN comment I ever made, but nobody cared to elaborate why my line of thinking is unethical or why i am reading the data wrong. That's slightly disappointing.

Exactly, I'm baffled by this as well. It's a simple and clear way to save many more lives and nobody even discussing this. It's also much easier to administer since you don't need to coordinate second visit. I don't understand why it wasn't part of the trial (some portion of people just get 1 dose)

Canada is the slowest among G7.

vaccines administered per 100 people: • Canada at 0.14 • United States at 0.59 • United Kingdom at 1.18 • Bahrain at 3.23 • Israel at 4.37

where did you find these figures? I've been trying to find some reliable source to keep an eye on it and I haven't been very successful so far

Check this out: https://ourworldindata.org/covid-vaccinations

The sources they use to compile the data are listed here: https://github.com/owid/covid-19-data/blob/master/public/dat...

Wish I could find estimates showing “At current / projected vaccination rates it will take X years to vaccinate 90% of the population”, per country. I heard on TWiV (This Week in Virology podcast) for the US it was ~6.6 years and I believe it was based on < 90% of pop.

_Current_ rates are likely not worth worrying about, not this early on. Rates should be expected to increase fairly dramatically in the next weeks.

I hope so. But, frankly, most national and local responses to this outbreak have been subpar. I do not take it for granted that we (the US) can pull off this major logistical challenge smoothly or on an aggressive timeline.

New vaccines will likely be approved over the next few weeks (particularly the Oxford one), greatly increasing supply, and the kinks around distribution and administration will be slowly worked out (the Oxford one will also reduce the logistical challenges there; no special transport or storage needed). I'd be very surprised if rates don't go up considerably within a month.

Is that likely to get approved in the US? My understanding is that the trial was not good enough to pass the FDA bar.

Ah, looks like they may hold off til Feb. for US approval. (Results expected from another trial by then)

Here’s why I worry: https://twitter.com/ashishkjha/status/1343768397084053505?s=...

Thanks, that's awesome

this one is also good https://www.bloomberg.com/graphics/covid-vaccine-tracker-glo...

Huh? The G7 is Canada (0.14), France (no data), Germany (0.05), Italy (no data), Japan (no data), the United Kingdom (1.18) and the United States (0.59).

We only have data on 4/7 in the G7 (at least at [1] and in your comment) and Canada isn't even last amongst those 4.

[1] https://ourworldindata.org/covid-vaccinations

You know that the G7 also includes France, Germany, Italy, and Japan, right? Which are all below Canada's vaccination level so far.

That and I'd think vaccination rates depends not only on how easy a country's population can afford such, but also, and perhaps more so, the level of anxiety (both, fear of severe COVID-19 as well as side effects of vaccine).

Honestly no-one should be reading _too_ much into numbers, especially all-time numbers, just yet. The UK approved early, so it has a head start. Most EU countries only started in the last few days. The US and Canada are somewhere in between.

Daily numbers will start to become more useful in the next couple of weeks.

In real terms though that head start could be significant given how quickly things can get out of control with this virus. The U.K. has been vaccinating for nearly a full month now which accounts for 750,000 of the most vulnerable people. That’s a good head start.

These creepy vaccines are far more scary than any virus that causes flu symptoms

This seems very Canada specific, is any other country doing it like this? I'm fairly sure neither the US nor most european countries are doing this (I mean, Canada seems to be the only place that are reserving the second doses)

Edit: This article is clearly being upvoted based on title alone

I’ve seen multiple experts in the media mentioning this idea in Belgium. Not sure if it is really considered by the government though.

The initial vaccination statistics I saw over the weekend suggested that the US/UK have both far outpaced the Canadian vaccination rate (adjusted for population). I don't want to cite specific numbers as I don't remember what they were, but it was a pretty large gap among countries that had approved the vaccines.

Tony Blair suggested it

That famous epidemiologist.

I had read that the USA is/was initially ensuring there were two doses onsite for each person vaccinated, until they felt the supply chain had proven to be consistent. I'm having trouble sourcing that though.

I saw that too and it makes sense. Begin with the most conservative approach and transition to just in time as confidence builds.

Yup, you don't want to send out a lot and having doses going bad because they were sitting around too long or some other issue.

The US is DEFINITELY doing this. Each time a first dose goes out of the Pfizer warehouse in Michigan, the second dose for it is set aside. Not only that, but they literally set aside ANOTHER 500,000 doses "just in case". They seem to be prioritizing smoothness of rollout and equity across states over immunizing the maximum number of people possible.

Whelp, guess I'm wrong on this one. That is extremely disappointing

Yup. It's stupid. You shouldn't hold back 0, but >50% is too much.

What happens when people start missing their second dose because of insufficient supplies?

- Endless stories of incompetence in the media

- Lack of clarity in need for second dose

- Infections in suboptimally vaccinated people draw question towards efficacy of vaccine.

These are all going to have deleterious effects on future messaging and policy decisions. There's no way to survive this situation with your reputation intact as there are loudmouths on both sides, at least try to salvage the process restore confidence in the underlying medical science.

I've been in a situation where medical policy directly contravenes with the health of a dearly loved individual. We lost that battle and the lives of myself and my children will be forever changed. However, I can rationally see the merit of this approach.

Ontario was the only province to start with this strategy. Retired general Rick Hillier, chair of Ontario's vaccine distribution task force, discussed the strategy in a CBC interview on Dec 14 [1] just before the rollout.

[1] https://youtu.be/PSFX5qFVzhw?t=369

As many have been saying, optimizing to minimize case count has been where the entire handling of the pandemic has gone wrong. Let's not repeat this mistake yet again.

We need to optimize to minimize long term effects: death, disabilities, mental health, etc. These are things that actually matter.

What are you advocating for here?

It seems to me like case count and most long-term effects are linked. If we reduce case count sufficiently that we can stop quarantining, that improves mental health, and case count should be pretty strongly correlated with death and long-term effects.

In the case of rolling out vaccination, are you suggesting we do something differently? Is there any difference between optimizing for minimizing cases and minimizing those other things in a vaccine rollout? In my mind, to optimize for any of those things, you'd roll out vaccines asap.

> We're sorry but we are unable to validate your access to this article. Please adjust your browser settings or turn off any extensions that may be causing this issue.

Hard pass.

But we're supposed to prolong the temporary overreach of government power! /s

In practice it's likely that we'll do both. That is, all the doses will be used, and none reserved, as they arrive according to a priority schedule (ICU staff, care home staff, care home residents first, etc).

Then when it's time for the second dose, the high priority targets will get their second dose immediately as new doses arrive while lower priority targets may have their second dose delayed as a first dose is given to those even further down the priority list.

The result being that those at highest risk get the double dose ASAP while the twenty year olds get their first dose at the expense of the forty year olds having to wait a little longer for their second dose. This is because even a first dose while not fully protective is protective against serious covid-19 outcomes which is a big deal in itself.

According to this scheme twenty year olds will be getting their second dose sometime in their thirties. :)

All things being equal, perhaps. But first doses are medical professionals and other critical workers. Guaranteeing they will be here for the long haul is another worthy goal.

I'm not sure 'spray and pray' is a well-thought-out approach. Its perhaps another internet pundit who can take one number and divide by another number and become an instant expert.

The proposal is not to start vaccinating random people off the street.

It is to vaccinate twice as many high priority people.

This is not what the article is about at all. It is saying we should not reserve the second dose in the freezer for that patient and should instead immediately distribute it to at-risk populations. Then we should trust the supply chain will come through for the second dose and that a 1-2 week delay in case it did not would not overly harm immunity. It is perhaps a very sensible suggestion that would depend on the supply chain quality.

I think we should all make efforts to tone down our rhetoric and not tilt at windmills.

Sure it is.

Again, if there should be some issue with the supply chain, then the first dose could be 'wasted'. For hospital personnel etc that would be a bad scenario.

We're not talking millions/billions of doses. We're talking about what are really very small numbers. It may be worth taking some very small hit in herd immunity, for an assured continuity of care.

> Again, if there should be some issue with the supply chain, then the first dose could be 'wasted'.

It's not "wasted". As noted in the article, the first dose alone is ~50% effective; the second dose is required to boost that to ~93%. So, it's a tradeoff between vaccinating n people at 93% effectiveness versus 2n people at 50% effectiveness.

Is there an actual serious claim that the first dose would be wasted?

This claim is being made despite CLEAR CDC guidance on how to handle delayed vaccination doses?

I'm serious - where is the citation for the claim you are making that the first dose is wasted if the second dose is not perfectly on schedule.

The reason this claim is likely a lie is that millions / billions of vaccines are given out. Every school child in America is getting vaccines. With very narrow exceptions, if you are off the exact timing (ie, you get a 12 month vaccine a bit later, you get the second shot 14 days later) which by the way happens ALL THE TIME because doctors and parents are busy etc - there is no evidence that the first shot is wasted or that you need to restart vaccination.

So please provide a citation for the "first shot wasted" claim.

Provide citation yourself for this case. Nobody can, there is no data to cite. The only data that we have is that getting both doses on schedule works. Everything else is speculation.

The issue is not who gets the vaccine, but whether to hold the second doses in reserve or distribute them as first doses and use later-arriving doses as second doses. The former has the obvious advantage of vaccinating more people quickly, but runs the risk that if the second doses don’t arrive on time the effects of the first dose will wear off.

Edit: I originally used the phrase “back to square one” but I agree with replies that it’s the wrong phrase.

Why is one "back to square one"? Pfizer vaccine schedule is the second dose at 21 days after the first dose, but as I understand, in theory, the second dose after a month, or even two months, should work equally well (except obviously reduced protection in between).

Sure it is about who gets it. Saved for 2nd doses for critical workers, or distributed to those further down the list now.

The first dose is already 50% effective, so you are not back to square one. The second dose also doesn't necessarily need to arrive precisely on time.

Reply on news.ycombinator.com