alt Hacker NewsThe two points you raise with regards to why vaccine or similar studies may be treat special, it doesn't replace the loss of data when a double blind study with a control or make estimates based on modelling indicate anything more than correlation.
We may broadly agree that submitting a control group to a placebo treatment for a particular disease is immoral, but that doesn't mean such a study isn't necessary to prove out the efficacy or safety of the treatment. As for modelling, for example trying to estimate FN and FP, it can only ever indicate correlation at best and will never indicate likely causation.
But it's not. You can do an RCT of the new treatment vs the old treatment. You won't get a direct measure of its absolute efficacy but you will know if it's superior/ non-inferior to the best known thing. And then you can use observational techniques to estimate the absolute values. That's exactly what you would do if you wanted to develop, say, a new flu vaccine that you thought would outperform current vaccines. You get the most important information: whether or not we should switch to the new one.
If you have a new vaccine for a disease for which there is no existing vaccine you do a standard placebo controlled RCT which gives you a direct, high quality measurement of efficacy and side effects.