Hacker News Reader

Pfizer has the highest efficacy endpoint (95% at preventing symptomatic infection, which would theoretically also prevent spread and symptoms at higher percentages) with equal efficacy across age groups.

Moderna is 95% at preventing symptoms (100% for serious symptoms in their trial but insufficient sample size). They also reported slightly lower efficacy for the elderly.

AstraZeneca is 62% for preventing symptoms (unknown for infection and spread).

So it's not quite a trade-off of spread prevention vs symptom prevention, because Pfizer claims to do both significantly better.

edit: reading the data in more detail, all 3 don't have sufficiently significant data to claim anything other than symptoms. So preventing spread is just PR-hype.

> Pfizer has the highest efficacy endpoint (95% at preventing infection

Cite, please. Pfizer's primary outcome measure was symptomatic infection. Only Oxford/AZ in the UK carried out weekly PCR tests to check for asymptomatic infections.

Looks like it was edited? Yea, only symptomatic infections were tested by Pfizer iirc

Reading the journal paper from AstraZeneca[1]. I don't think AZ can claim anything about preventing asymptomatic infections. You're right in that they're the only ones with that data, but that data is practically identical between control (37/2760) and vaccinated (34/2751) groups.

Side note, the more I read about the AZ trials in detail, the more it becomes obvious how messy their trial was compared to the Pfizer/Moderna ones. They have so many more methodologies and outcomes and much smaller samples.

[1] https://www.thelancet.com/journals/lancet/article/PIIS0140-6...

They did find fewer asymptomatic infections in the half+full dose arm but the N is not large enough.

The other trials simply didn't collect data about impact on transmission. So there isn't data to support a statement in either direction.

Typically, high levels of antibodies will suppress infection and thus transmission. But there isn't data to make a conclusive statement about it.

> Unlike other vaccines being trialled, the Oxford team had been taking weekly swabs from all volunteers to check whether they were infected but showing no symptoms.

FTR, this only happened for one of the trials (the AZ results are actually from several trials combined) in England and Wales. In Brazil, for example, this was not done.

Presumably they also counted the numbers of cases with symptoms in control and experiment? I assume the ratio there was also not as favourable as in the mRNA vaccines?

> no-one who got the vaccine was hospitalised or got seriously ill due to Covid. Whereas in the control group there were 10 serious cases and one death.

I can only imagine what it must be like for these researchers. You give some people a dummy and let them get seriously ill and one even dies.

No idea what else we could possibly do, this is probably just how this must be for the greater good, but wow must that feel horrible for everyone involved. You have to tell the family "Sorry, she was in the control group!" Peak 2020 moment for everyone involved right there.

Double blind trials are the standard for establishing efficacy. Unfortunately, there is no ethical way of getting around the problem you mentioned.

We should have been doing challenge trials from the start. Hundreds of thousands of lives if not more could have been saved, and people of sound mind could have decided for themselves if they want to participate and accept the risk. I know I would have.

I wonder if more authoritarian governments like China or Russia would be willing to do challenge trials.

Outside of ethical concerns, it comes with other complications, like how to design infecting the challengers so that it would be representative of real world infection/viral load.