Unfortunately, human research isn't as simple - you actually have to worry about the health consequences as well as the benefits. But correlations can be very meaningful. For instance, cross-cultural studies w/r to diets can be illuminating even as there are no experiments being run. Still, I'd place my bets, and my life, on fresh foods - high in anti-oxidants - over processed ones. The beauty is it ain't rocket science. And better, good diets can control and even inhibit pathologies (diabetes, heart disease, etc.).