Peter Attia· MD
if you had a hazard ratio of 0.5 but your confidence interval was 0.4 to 0.6 or 0.7 or even up to 0.9 you would say indeed that is not at the level of 95 percent that is confident
Loading…
The headline is broadly defensible, but the qualifications matter. Effect sizes vary by population, the strongest claims rest on shorter trials, and credible voices push back on how it's typically framed.
if you had a hazard ratio of 0.5 but your confidence interval was 0.4 to 0.6 or 0.7 or even up to 0.9 you would say indeed that is not at the level of 95 percent that is confident
Every Sunday: the week’s new conflicts and verdict changes — and nothing else.
Native comments, Twitter mentions, and Reddit threads about this claim — surfaced together so the conversation isn't fragmented across platforms.
Bookmarking — the dossier-vs-overview split is the right call. Most of the time I want overview; sometimes I want receipts.
Would love a "what would change this verdict" RSS feed. Sign me up if it exists.
you gave an example a second ago Bob you said your hazard ratio was what uh Hazard ratio is 0.5 0.5 okay with a confidence interval of 0.4 to 1.2 okay so that would not be significant