The accuracy of this test is nowhere nearly good enough to do population-wide screening. The clinical setting for this test is memory clinics in which Alzheimers is already relatively highly likely differentially, and even there you're going to get a surprising number of false positives.

(There's enough info in the supplemental link on this page to have an LLM do the Bayes math for you.)

Off-topic: I only came to this thread because I thought "I bet some comment will manage to somehow include LLMs in the discussion".

Looks like my prior was not too bad :)