Originally Posted by
Carbonfiberboy
I saw a study in JAMA, maybe a year ago, which analyzed as many nutritional studies as they could find for the past several years. They concluded that about 80% (I don't remember the exact number, but like 78%-83%) contained researcher bias. I never believe anything that the author is making money off of, at least not without multiple confirmations. So far, I've seen one book which wasn't obviously biased, but it wasn't worth reading anyway, just the usual good nutrition advice.
One major problem is that NIH funding is focused on diseases, rather than sub-clinical health effects of diet in the general population. As a result, a lot diet research, including some of the best, has been funded by food industries, tainting it to varying degrees. This is, hopefully, changing with the realization that symptomatic disorders (obesity, T2DM, cardiovascular disease) are late stages of a condition beginning years or decades earlier.