There is double blind studies that tests efficacy of new vaccines for diseases that already had an approved vaccine. And those studies do find differences in efficacy (just because the old vaccine was 96% effective doesn't mean the new one can't do better) as well as in occurrence rate of side effects.

There have even been studies where the vaccine that you get in the control arm of the study is the vaccine against Meningitis and then after a year or two they do give you what ever vaccine you had not received yet. (Sorry, forgot against which disease the vaccine under test was protecting.)

And of course there is double blind studies where you get three injections, one of which is saline, two are active. They are given a few weeks apart, you don't know the order and they check antibody level in the blood every week.

WTF?

https://www.google.com/search?client=safari&rls=en&biw=2326&...

Look up "efficacy" it has a specific medical meaning. Those tests are not about efficacy.

Scientific method require each person be equally exposed to the virus under strict lab conditions.

Scientific method uses statistics and can make reasonable assumption that for large enough randomised groups the collective exposure is the same without strict lab conditions. From the group size and number of infections, you can calculate the confidence of the result you find.

Keep in mind that even under strict lab conditions your results would never be 100% certain.