Marty Makary, Professor of Health Policy at Johns Hopkins, as quoted by Alex Tabarrok:
Ironically, those in the Oxford-AstraZeneca trial who inadvertently received half the initial vaccine dose had lower infection rates
Makary and Tabarrok’s main point (with which I fully agree) is that it’s criminally stupid for the FDA not to approve the A-Z vaccine immediately — and their main argument would stand with or without the observation about infection rates.
But I’m quoting the same observation for an entirely different reason: To point out that sometimes you need economics to explain the medical data.
In particular: Half-dosed subjects will generally have fewer side effects. Subjects with fewer side effects will think it more likely that they’ve gotten the placebo. Subjects who think they’ve gotten the placebo are going to continue taking more precautions with masks, social distancing, etc. Therefore it’s entirely plausible that half-dosed subjects will have lower infection rates.
Thanks to Romans Pancs for pointing me in this direction, and reminding me of the Thanksgiving puzzle that I posted here.
It’s also entirely plausible that half the dose provides substantial protection.
That seems a plausible explanation on the surface. However, there are others. If this is the explanation we have to assume that those in the vaccine arm behaved a lot differently than those in the control control arm. Also, the second dose was full strength and 3 weeks later or thereabouts. Most of the infections came after this. I don’t know the side effect to the different doses. It would have to be nearly all in reaction to the first dose, which is possible, but could be checked to verify if this mechanism is plausible.
The interesting discussion is about the delay between 1st and second doses. UK is going to gove as many as possible the first dose, leaving it much longer than in the trial to get the second. The idea is that with cases raging it is more important to prevent serious illness and death in as many as possible, which the first dose seems to do. The risk is the second dose might not work as well, plus we increase the risk of resistant variant breakout. Real risks. I think the risk is probably worth it, but reasonable people could certainly disagree on that.
Even better, the FDA could have approved these vaccines six months ago, and saved 100,000 lives.
They would have looked a bit silly if the trials then showed they were not safe. I think given the seriousness of the situation more short cuts should have been taken. However, we live in a political world and any fallout from a failed unsafe vaccine rollout would likely have very serious consequences for vaccines for decades. It was probably best overall to do the usual trials as quickly as possible.