My dad, I and my husband were fully (?) vaccinated; then dad went to a medical clinic (vaccinated people only, all wearing masks), he got infected, I caught covid from him and passed it to my husband. Dad was very ill with covid, stayed in hospital, needed oxygen and antivirals. Now of course someone will come up with one of these canned responsed:
well, maybe one of you was vaccinated too many months before, so you're irrelevant for our statistics (but how many months are too many? do we need jabs every 3 months? 2 months? 1 month?)
well, maybe one of you was just vaccinated and this was a ”breakthrough” infection, so you're irrelevant for our statistics (but when does this vaccine start being effective? 2 weeks? 1 month?)
well, all of you are still alive, so it must be the vaccine that saved you; maybe without jabs you'd be dead! (Because this is how science and public policies works, based on ”what if” and ”maybe”.)
Evidence based medicine doesn't rely on anecdotes, and yes all science is caveated with a level of uncertainty @Terzani
We know that vaccinated people are less likely to get infected, and have better outcomes. Thus it is likely your Dad would have been worse off had he not be vaccinated, but obviously no one can say for sure in an individual case unless you had a paralell universe where he was unvaccinated to compare to.
As epidemiologists know this is the issue they're trying to get around - inferring causality when you are not able to repeat a study in the parallel universe when the event did not occur (the "counterfactual").
I know someone who died despite having chemotherapy for a specific type of breast cancer. That doesn't mean if I developed the same type of cancer I wouldn't have chemo - because evidence from large samples of people are used to recommend this course of treatment.