Thank you for your well-reasoned comment. First, I cited studies correlating protection against dementia from other vaccines based on prior observational studies, which are prone to the healthy vaccinee bias.

However, the observational study by Eyting et al. managed to minimize the healthy vaccinee bias by capitalizing on the naturally randomized situation surrounding birth date-based eligibility for shingles vaccination.

Interestingly, Eyting et al. also showed that the vaccine-eligible group did not seek any more vaccinations than the vaccine-ineligible group. This suggests that dementia protection may be specific to herpes zoster/shingles vaccination.

But as you have mentioned, there might be other interaction vaccine-vaccine interaction effects influencing dementia risk that Eyting et al. did not study.