{
"$type": "site.standard.document",
"bskyPostRef": {
"cid": "bafyreiahnavc6k5tcqqptrp2v7ikzeoyqs7fuerqqhuhnbmoh7f4okwati",
"uri": "at://did:plc:wwyqal4cnqhuwyacdj7rqq3n/app.bsky.feed.post/3meszhtz2fe52"
},
"path": "/t/statin-related-side-effects-the-recent-lancet-publication-is-biased-toward-false-negatives/28635#post_9",
"publishedAt": "2026-02-13T18:23:55.000Z",
"site": "https://discourse.datamethods.org",
"tags": [
"well-understood"
],
"textContent": "f2harrell:\n\n> I’ve always thought that absolute risks should be emphasized much more.\n\nProviding absolute risks differences would help contextualize AE info. Another form of contextualization (not sure a Fundamentalist Bayesian would agree tho’) would be _substantive knowledge about mechanism_ : is this rare AE well-understood, or is it (so far) just a data artifact lacking a plausible mechanistic explanation?",
"title": "Statin-related side effects: the recent Lancet publication is biased toward false-negatives"
}