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.
The one large randomized controlled trial that was done on 9,000 people, not on 90 people or 50 people or 30 people, which a lot of these studies are, but on 9,000 people that were randomized in a psychiatric hospital, it would be unethical to do today. It was done by Ancel Keys. It was the Minnesota coronary experiment, funded by the NIH, where they basically gave half the group butter and half the group corn oil.
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.