Don't take anything said about vaccine effectiveness seriously unless
1) samples are differentiated by lifestyle, by diet, and by other activities thought to strengthen immune systems
It is ridiculous to ignore this point, given that the effect of something (vaccination) on the strength of immune systems is precisely the topic in hand.
Let's say you're considering how good a Super Expert Duper (SED) doorlock is, meaning how likely it is to stop intruders getting into your house.
If you have numbers telling you how many houses with SED doorlocks have been broken into, and how many houses without SED doorlocks have, then so what?
That information doesn't tell you how effective SED doorlocks are. The doorlocks could be fake, they could lead to more break-ins, they could lead to fewer.
You need to know about kinds of area, whether people are at home all day, whether they frequently go out leaving other doors or windows open, how close the houses are to other houses, and so on. You need this information both for homes without SED locks and for homes with them.
There are many things people can do to improve home security other than fitting a Super Expert Duper doorlock - as anybody who cares about home security, who has a bit of sense, and who isn't a rep for the Super Duper Company is aware.
There are also many things that real people do (or omit to do) in the real world that are awful for home security - like live in high-crime areas for example.
The same is true about immune systems. Age is not the only variable that should be considered.
Physical exercise, vitamins B12, C, and D, iron, a vegetarian diet, and so on - these tend to strengthen the immune system. So compare vaccine effectiveness relative to the effectiveness of THOSE.
2) there is clarity regarding the level of confidence that should be assigned to estimates of total population sizes - for example when statistics are cited such as "x% of y are z", ask "how did you arrive at y?"
One can boil this down to a simple question: what is the size of the unvaccinated adult population in Britain, estimated within a 95% confidence interval?
3) you can explain reasons for your level of confidence that government statistics in Britain are much more reliable than they were in say the USSR.
I simple terms, the published figures are deliberate propaganda; don't believe in them by default.
All of the above is why trials balance the groups and controls.
Properly designed trials (such as many done) almost completely remove all those potential confounding variables.
