We need more randomized trials in nutrition–preferably with negative results (ajcn.nutrition.org)