Regarding the “caveman” trend:
Question!: Just read your piece and scanned the article in the Times (sorry but it is way too NYT for me). My question is, if we are going by 'ancestral diets' shouldn't be different for many ethnic groups? What do you do if you are a mix like me? Do I eat dairy because I like it and dairy gave my pastoralist ancestors a big leg up or do I eat tomatoes and potatoes because of my Indian ancestry? What about fish? I have a lot of Norwegian, should I eat lots of rotted shark, salt cod, and salmon, what about whale or seal (very greasy)? I think that some groups might do better under different diets but how would you know what is best for you?
I suppose it’s the same as for disease risk alleles, like type 2 diabetes. What difference does it make to be half Pima? Are the genetic influences continuous or Mendelian?
Of course we have no clue.
But on the general question, it seems that SW American Indians do better on more “traditional” diets; N Europeans on average do better on milk than S Europeans, and so on. But then diet is such a broad subject – so many distinct foods enter in, and it takes really drastic diet differences to find any large difference in health between different groups, and very large sets of uniform individuals to find statistical significance.