[x86ARMsRace]

As a Canadian (read: lab rat) I'm hoping this forced experiment I'm participating in, run by someone with no scientific background, ends well also. Thousands of lives are depending on what, at this point, amounts to a gamble on minimal evidence.

[jeromegv]

This was not a personal decision from the prime minister, this is from the National Advisory Committee on Immunization. You can disagree of course, but let's not lie and pretend that this is just Trudeau that woke up one morning and decided to do that, lots of people are behind that decision and they also had the benefit of insight looking at what the UK had been doing.

[smnrchrds]

Yeah. I count 13 doctors and professors out of a team of 14 people. "someone with no scientific background" is not an accurate description of the decision-makers.

https://www.canada.ca/en/public-health/services/immunization...

[sunstone]

Not to mention that the Chief Medical Officer of British Columbia, Dr. Bonnie Henry, came to the same conclusion even before the national level bodies had decided. Dr. Henry implemented the strategy by stopping second vaccinations so more people could get first vaccinations.

[na85]

In Bonnie we trust.

[theptip]

Let me attempt to give a defense of "first shots first" as someone who initially wrote it off as foolish, and later came round to it.

In an ideal world, you'd do a few clinical trials and figure out the optimal dose regime and timing. (I'm not really sure why they didn't run 5 parallel trials with different timings to begin with).

In the world we inhabit, alas we didn't have those studies. But that doesn't mean we have no reason to believe that the first shot is protective! You can still look at how vaccines in general work, and our basic understanding of immunology. In general, for vaccines with multiple shots, the model seems to be that the first shot gets you >50% of the protection, and the second shot mainly extends the protection over time. This might not be how the COVID vaccines work, but your priors should be fairly strong that this is what the clinical trials will ultimately show, because that's what they have shown for a bunch of different vaccines with different mechanisms.

If you're in an ideal world where inaction doesn't cost you anything, sure, do the studies.

In the world we inhabit, thousands were dying every day, and so inaction is morally horrifying. Giving a pair of doses as first shots to two >65yo results in far fewer people dead than giving that pair of doses as a full course to one 65yo. The expected value is pretty clear on this one, even though the clinical trials were not.

You should probably have started to feel tentatively confident that the first-dose-first strategy was sound circa mid-Feb, when it the UK's death rate had been falling rapidly for about two weeks: https://lh4.googleusercontent.com/GGlgMxEKIoaxyuZbuHjjEBH2er... -- while it's hard to disentangle lockdown effects, EU vs. UK was mainly a difference between "locked down with vaccines" and "locked down without vaccines", at risk of oversimplifying.

If I had to summarize in a sentence: use Bayesian reasoning rather than applying unjustifiably strict error bars on your decision process.

[saurik]

> Giving a pair of doses as first shots to two >65yo results in far fewer people dead than giving that pair of doses as a full course to one 65yo.

In the US, though, I don't think we have ever been forced between these two scenarios: we have always had enough vaccine doses for people in our high risk classes. The question thereby is more whether it is better to have all the essential workers and everyone over 65 years old along with maybe third of everyone else with a single dose or to have the first two classes of people fully dosed... and "I dunno", right? ;P

[theptip]

> I don't think we have ever been forced between these two scenarios

I'm not sure; January->April the US was constrained by vaccine supply. The initial rollout to at-risk groups was not a case of "we got a shipment of 100m doses, now we need to figure out how to use them", it was more "we're manufacturing 1-2m doses per day and trying to ramp that up". It took months to go through the gradually broadening risk tiers. (e.g. see https://investors.modernatx.com/news-releases/news-release-d..., which says "deliver 100m vaccines by end of March, 100m more by end of May").

I believe the US was keeping shots back in favor of using them later as second doses, rather than giving them immediately as first doses, though I don't have a citation for that to hand. If that's right, the US could have ~doubled its rate of vaccination, i.e. got to the April 15 "all age group open vaccination" milestone in 2 months instead of 4 months, with first doses only administered to that population. At the very least, we were giving second doses in Feb (21-28 days after first doses in Jan) that could have been given as first doses to the higher-risk groups we were still prioritizing through April.

There's a follow-up question on which I have not run the numbers -- are there any X,Y pairs for which you'd rather hold back a shot to give as a second dose later for a >X year old, instead of giving it immediately as a first dose to a <Y year old. Interesting question, I don't think that's what you were getting at though.

[YZF]

As a fellow Canadian I'm happy that finally we get some common sense in this pandemic. If we hadn't taken this approach we'd still be heading up on case counts instead of heading down, increasing the vaccinated base is what has turned the tide on the third wave and is what's going to end the pandemic here months sooner than it would have ended otherwise.

We don't have time to RCT test all scenarios. That's how life works. The long term immunity as a function of delta between the first and second shot has not been tested at all. J&J vaccine is one shot mostly because it can only be one shot yet it still works. There are other aspects that have not been tested in the trials. We need to make reasonable decisions under uncertainty in an ever changing landscape.

[ahmedalsudani]

There is plenty of evidence that the first shot reduces the risk of the worst symptoms quite significantly.

[x86ARMsRace]

> reduces the risk of the worst symptoms quite significantly.

Right, reduces, but the full dose is needed to render the virus mostly harmless. My partner's coworker has Covid and one does of the vaccine, they are still severely ill.

[smnrchrds]

First of all, when we read or hear about severe illnesses due to COVID in the news and articles, what they mean is hospitalization, ICU, and death. Are your coworkers in the hospital? If so, my thoughts and prayers go to them. I hope they make it.

Second of all, the choice was never between giving two doses to 16 million people (as of today) or giving one dose to 16 million people. The choice was giving two doses to 8 million people or one dose to 16 million people. The total death toll from the pandemic would be lower the sooner we get to the point of herd immunity. Giving everyone one dose gets us there sooner and ultimately saves lives.

[saurik]

FWIW, in the United States, it isn't clear we are ever going to herd immunity... like we are at the point where the New York Times is just reporting on the idea that we've lost that battle. So I can feel sympathy for the strategy of getting our essential workers and old people--at lease the ones willing to take a vaccine--double-doses before moving on to people who are less likely do die from the disease.

[pcbro141]

Do you happen to know roughly how long after getting vaccinated they caught covid?

It takes ~14 days after the first dose to get the ~80-90% protection (Pfizer/Moderna).

[__xyz]

Unfortunately, I think this is not true for the latest mutations that appeared within the last months (e.g. B.1.1.7 from the UK).

Please correct me in case you have different information