The vaccine trials concentrated on symptomatic infection, and those are the results you see quoted. Some also had sub-trials with repeat testing of the same people or randomised testing of a subset each week. They all looked at bloodwork to get antibody data (for vaccine and for infection) but I've not seen those results published.None of the vaccines are sterilising (totally prevent infection). The effectiveness stats aren't always very comparable because some trials only tested for covid infection when a trial participant presents symptoms which means asymptomtic infections wouldn't be found so it is a measure of how well the vaccine prevents symptomatic and serious disease. If a study tests everyone in the trial for covid then the effectiveness figure is a representation of how well the vaccine reduces any infection.
@Pogue Mahone @Tony Babangida @Volumiza @africanspur - is that right? I also have a recollection (quite possibly wrong) that Moderna and Pfisher were in the former group (didn't test for asymptomatic infections) and AZ/Oxford were in the later group (tested all participants). If this is the case then that might explain some of the gap in effectiveness between AZ and the mRNA vaccines. No idea about the other vaccine's trial methods.
AZ did the biggest of those weekly retest trials but the initial data was erratic, and covered multiple dosing patterns, averaging out at about 20% efficacy against asymptomatic infection.
AZ https://www.cidrap.umn.edu/news-per...how-astrazeneca-covid-vaccine-has-90-efficacy
The great unknown with the vaccines is whether the ones who catch covid with/without symptoms are less infectious than the uncaccinated who catch it. Early suggestions from the Israel and UK mass rollouts are that cases are milder and the infected less likely to spread the virus - but that's all about averages and big statistical models, so it's a wait and see.