Thread by @VikiLovesFACS

@RealNormalPod @lisadunn1978 Great question! What you've just described there is called a "challenge study" and is sometimes done, eg for malaria vaccines. But in the case of COVID, since it's potentially fatal and there's no cure it was considered unethical. 1/

@RealNormalPod @lisadunn1978 In the trials, what they did was vaccinate half the participants and give the other half a placebo (either salt water or a different vaccine). Then they regularly tested everyone to see if the people who got the vaccine were less likely to get a positive test than the rest. 2/

@RealNormalPod @lisadunn1978 In fact, not all the trials did this. AZ did and Moderna sorta did but Pfizer only tested people when they got sick. This is why we spent some time when the vaccines were new going "we know it stops you getting sick, but we don't know if it stops you getting infected." 3/

@RealNormalPod @lisadunn1978 Anyway, we do now know that all the vaccines reduce the chance of getting infected, and we can do this with real-world testing data. The best approach to this is to look at people like me who take a weekly test for work whether they are sick or not... 4/

@RealNormalPod @lisadunn1978 ... and then see if the ones who are vaccinated are less likely to get a positive test than the ones who are not. 5/

@RealNormalPod @lisadunn1978 The PHE data I posted above took a slightly different approach. They looked at tests that people had requested because they had symptoms. They then split the tests into those that came back positive and those that came back negative and asked... 6/

@RealNormalPod @lisadunn1978 "What % of people are vaccinated in each of these groups?"

You'll notice that that doesn't exactly tell us what % of vaccinated (compared to unvaccinated) ppl get COVID. But there's a statistical theory you can use to rearrange the equation and work it out. 7/7