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.
epidemiology Our survey based studies there is no experiment there's no control group they'll take a group of people and they'll survey them to look for health conditions how much heart disease how much cancer how much diabetes do you have and they'll give them a either follow them moving forward or follow or do a survey based on their recall retrospectively and ask them what they ate
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.
around epidemiological evidence being used to formulate hypotheses which need to be tested and in more controlled experimental studies