Randomized controlled diet experiments over the lifetime of a human are pretty difficult. So they make do with what data they can get. Putting that data togeth

Randomized controlled diet experiments over the lifetime of a human are pretty difficult. So they make do with what data they can get. Putting that data together to form a recommendation is incredibly difficult. If we could just get people not to imagine that it's trivial ("eat no isolucene and live forever"), we'd be several steps ahead of where we are.