Eric Topol

Yeah, and that brings us to the natural experiments.

I just wrote about recently the one with shingles, which there's two big natural experiments to suggest that shingles might reduce the risk of Alzheimer's, an added benefit beyond the shingles that was not anticipated.

Your thoughts about natural experiments, because here you're getting a much different type of population assessment.

Again, not at the individual level, but not necessarily restricted by some potentially skewed enrollment criteria.

Yeah, no, I think it's really valuable.

And now I think increasingly given priority, if you can find these natural experiments, they're not always so abundant to use, you know, to extrapolate from, but when they are, they're phenomenal.

The causation correlations,

is so big, the issue there.

I mean, Judea Pearl's Why book, and you give so many great examples throughout the book of proof.

I wonder if you could comment on that a bit more, because this is where associations...

are confused somehow or other with a direct effect.

And we unfortunately make these jumps all too frequently.

Perhaps it's the most common problem that's occurring in the way we interpret medical research data.

Yeah, I think this cause and effect is a very alluring concept to support proof, as you so nicely go through in the book.

But of course, one of the things that we use to help us is the biological mechanism, right?

So here you have, let's say, for example, you're trying to get a new drug approved by the Food and Drug Administration.

And the request is, well, we want two trials, randomized trials, independent.

We want to have p-values that are significant.

And we want to know the biological mechanism, ideally with a dose response of the drug.

But there are many drugs, as you review, that have no biological mechanism established.