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.
what we're looking at here are associations and correlation does not always mean causation sometimes correlation can be an indication of causation but usually ideally these studies generate hypotheses that are then tested in interventional studies unfortunately as you and I both know nutritional studies that are interventional are very rarely done
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.
look we are stuck with epidemiology we cannot draw causation from correlation but sometimes epidemiology is all we have and so we must acknowledge his limitations and use it
It's correlative and like blah, blah, blah. It doesn't mean anything, which is sort of an argument you could make about most nutritional studies because that's the very nature of how they're done, right?