So epidemiological studies can answer whether something isn't bad, but they can't answer whether it is bad. If you think eating red meat it bad, and you watch a

So epidemiological studies can answer whether something isn't bad, but they can't answer whether it is bad. If you think eating red meat it bad, and you watch a million people eat red meat or not, and there aren't adverse effects, it has given you pretty good evidence red meat isn't bad. However, if you see a difference in the red meat group, you still haven't shown red meat it bad. The problem is they get used incorrectly all the time, which is why I consider them nearly useless. I actually came to this more nuanced view after having conversations with Dr. Feinman about them. :) I think it's a completely valid question to ask why Feinman, Westman, et al, who all have financial ties to the their viewpoint, should be trusted. All scientists are financial vested in their results in some way, so you always have to be careful. (An aside: Dr. Feinman is an interesting case study, though: he was a biochemistry instructor at a medical school. He had no training in nutrition at all. Because of the biochemistry, he taught his students that fat is a great energy source and too many carbs are dangerous. At some point, he was told he wasn't allowed to teach that anymore, because it went against the nutrition orthodoxy (even though it followed the biochemistry), which was when he entered the nutrition arena.) Fortunately, we have better methods than check stubs to determine whether to accept something. I look at declarations as indicators of possible bias, but, in the end, a study is good or it is not good. It's unfortunate that it literally requires reading and understanding scientific studies if you want to figure this stuff out, because it is all so bogged down in politics and economics. It's even more unfortunate that many studies get a lot of press that fall squarely in the "not good" category. I've seen great studies come from researchers who set out to "prove" vegetarianism was the best diet. I've seen great studies from people paid by Atkins and by Ornish (a noted very low fat vegetarian proponent). Some of the worst studies I've seen have come out of Harvard, but even those aren't uniformly bad. The weird thing about nutrition is there is this huge disconnect between the science and the organization supposedly representing the science. The American Heart Association has studies that show saturated fat is not unhealthy, yet, as an organization, it continues its zeitgeist against it. The science is clear that sugar is absolutely horrible for diabetics, yet the ADA continues to tell diabetics it's fine to eat. Registered dietitians get continuing education credits from classes literally taught by McDonalds and CocaCola. I cannot say with complete confidence that it is because of their sponsors, but I do think, at a minimum, the influence of corporate sponsorship as a factor in resisting change is a reasonable hypothesis that fits the data better than any other hypothesis. So, is it possible that the ADA doesn't come out strongly for a low carb diet for Type 2 diabetics for a reason other than the fact they get most of their sponsorship money from companies that produce insulin and other diabetes drugs? Sure. A certain amount of it is inertia based on old, bad science, but I think the sponsorship money acts as a huge anchor, keeping that inertia in place.