"The risk/benefit of Covid vaccines is arguably most accurately measured by an all-cause mortality rate comparison of vaccinated against unvaccinated, since it not only avoids most confounders relating to case definition but also fulfils the WHO/CDC definition of "vaccine effectiveness" for mortality"
But has the obvious confounder of assuming the two populations are equal in all other ways other than being vaccinated. This is probably false.
There was a US dataset where it was shown the mortality from a variety of diseases was lower among the vaccinated. This isn't because the vaccine is protective against those diseases!
Measuring via all-cause mortality is not advisable unless your population dataset is very large (a whole country is certainly enough though).