Dr. Pascal Geldsetzer

Thank you so much for having me.

Well, I think the shingles vaccination program, as it was rolled out in several countries, provides this really unique opportunity in observational data to get at cause and effect rather than just correlation, right?

We usually always have this fundamental problem in cohort studies, in electronic health record data, medical claims data.

that we have to compare people who are very different, you know, people who go get vaccinated versus those who don't, people who are prescribed a certain medication versus those who don't.

And we know there are different health behaviors, preventive motivations that drive these decisions.

And we just have very, I mean, it's very hard to measure these concepts in the first place.

And the measures that we have of these things in electronic health record data are even poorer.

And so we try to adjust for these differences in our statistical analysis, but we never know whether we get there.

And so we're always left with the question, is this correlation or actual causation?

Correct.

Yes.

Yeah.

Yeah, absolutely.

And then they have a lower risk of dementia in the future, regardless of whether they get the vaccine or not.

Right.

But in the way the shingles vaccine was rolled out in a number of countries, we've got all of a sudden very different comparison groups.

In fact, comparison groups that are very similar to what we would have in a clinical trial, which is the gold standard, right, to get at cause and effect.

So they said in a number of countries, so for example, in the UK, if you had your 80th birthday just prior to the start date of the shingles vaccination program, you were ineligible and you remained ineligible for life.

Well, if you had it just after, you were eligible.

And so now we can compare in our data people whose only difference is a week or so in age and