All of this highlights the need for more funding towards nutrition science. It's absurdly hard to do this stuff (e.g., controlling what people eat for a year),

All of this highlights the need for more funding towards nutrition science. It's absurdly hard to do this stuff (e.g., controlling what people eat for a year), so it's expensive. But given how expensive things like diabetes are, I think it'd be societally efficient to drop $100M or something on few grand randomized controlled experiments. This[1] is the closest thing I'm aware of. You'd sorta hope governments would make this kind of thing a priority. [1] http://nusi.org/the-science/current-science-in-progress/#.U9hDvIBdX4J http://nusi.org/the-science/current-science-in-progress/#.U9...