Will you take an approved COVID-19 vaccine?

I've been safe too, took loads of precautions at work and avoided mingling with people as much as possible. Didn't even have my annual head-cold in February. I also know nobody who has died or became very ill with it.

I'm fully vaccinated now though. As remote as the chances of becoming seriously ill with it might have been, I still didn't fancy it.

@floorpie wrote

We would see this effect if the vaccines were most effective in preventing symptomatic infection in people with highly functioning immune systems, because those people would be filtered out of the sample from which you derive your skewed death rate.

Which is why, as a bunch of people have told you repeatedly, you have to use the vaccinated population as your denominator, not the symptomatic vaccinated population, in order to determine whether vaccines prevent death.

Even then there are other possible biases in the populations, like age, comorbidities, exposure, timing of vaccination etc. We would expect the take up of vaccines in young people to be biased towards the those greater at risk, e.g. people with underlying conditions, healthcare workers, older people. To take a trivial example, the under 50 vaccinated population is almost certainly older than the under 50 unvaccinated population.

Right, that's one possibility, albeit it's a guess that vaccines reduce hospitalisation rates of healthy people with highly functioning immune systems, such that we'd see it reflected in deaths, or conversely that healthy unvaccinated people are dying at significantly high rates. Maybe they are, but rates of comorbidities in all hospitalisations (incl pre vaccination) are extremely high.

90% of adults are vaccinated by at least one dose both here and in the UK. There is of course bias but it's miniscule. There definitely is a question raised by the UK data re: time since vaccination

And as I have said repeatedly, vaccines are proven to reduce symptoms, and it's likely that absolute deaths are lowered due to this. The questions are still raised, that Aristotle waved away by repeating "per capita". It's interesting to see boards.ie users deeming CFR to be unimportant in epidemiology and vaccine development, and that the only important metric is mortality per capita. Lol.

You do not know how important or unimportant it is, in this case. It is possible that CFR increases for vaccinated people without comorbidities relative to unvaccinated people without comorbidities, isn't it? You/I do not know. You make a good guess that the data is skewed due to healthy vaccinated people, but it's a guess.

Regardless, I'd never make/suggest a decision to get/not get a vaccine based on such data for someone at risk. If you're in a risk category then clearly vaccination will likely reduce your risk.

Didn't say that, the extent of this is unknown, and is not promising in trials that properly assess asymptomatic infection

Yes it has (See AstraZeneca trials: Efficacy of the ChAdOx1 nCoV-19 Covid-19 Vaccine against the B.1.351 Variant | NEJM)

Essentially no people died during phase 3 trials so far, so yeah we have to rely on less controlled data

Re-read it to improve your clarity

"In this trial, we found that two doses of the ChAdOx1 nCoV-19 vaccine had no efficacy against the B.1.351 variant in preventing mild-to-moderate Covid-19. There were no cases of hospitalization for severe Covid-19 observed in the study"

That isn't immune escape, it is reduced efficacy against mild to moderate Covid-19, but not immune escape. The vaccines remain effective against all known variants of SARS-COV2.

.

How can you put those 2 sentences in the same post that only has 4 sentences

This is a phase 2 trial by the manufacturer of the vaccine. It is decisive and will not proceed to phase 3.

There seems to be a lot of confusion in this thread about conclusions that can/can't be taken from controlled trials vs data collected in the wild.

That's the correct paper, read the protocol and collaborators

I'm not confused, it was run in collaboration with AstraZeneca and is published by the same authors that published the phase 3 trial of AZ for the original strain (here: Safety and efficacy of the ChAdOx1 nCoV-19 vaccine (AZD1222) against SARS-CoV-2: an interim analysis of four randomised controlled trials in Brazil, South Africa, and the UK - The Lancet)

Here is my attempt at recreating the New Statesman chart from a few pages back (with current Irish data):

The blue is confirmed cases and the red is deaths. Each row represents a weeks aggregated data from the start.

The blip near the bottom is from the HSE hacking incident (where no data was reported for a few weeks).

You can see it here: Irish Confirmed Cases v Deaths / Gordon Smith / Observable (observablehq.com)