Stuart Buck on Overcoming Bias posted this great article summarizing Gary Taubes's book, "Good Calories, Bad Calories".

Gary Taubes appears to be making a strong case that diet-health research has developed a strong bias in the past few decades, which led to evidence being ignored that (1) cholesterol and high-fat diets might not in fact be as bad for us as we've been told, and (2) processed carbohydrates might in fact be much worse for us than we've been told.

My suspicion is that these correlations might all boil down to abdominal fat. Abdominal fat is really bad for you - especially for men. The cells in abdominal fat convert testosterone to estrogen. Because the brain regulates testosterone production based on the amount of estrogen in blood, this causes testosterone production to lower, which further throws your hormone levels out of whack. The elevated estrogen then proceeds to cause all sorts of damage, increasing your risk of heart disease and cancer.

Stuart's article mentions this from Taubes:
[In] 1961, a conference of the Association of American Physicians included a presentation showing that in comparing heart disease patients in New Haven to a healthy population, the diseased patients were much more likely to have high triglycerides than high cholesterol, thus implicating high carbohydrate diets (which elevate triglycerides).
I would like to know if this result was controlled for the amount of abdominal fat.

My suspicion is that, in the end, as far as health is concerned, diet doesn't really matter. Just stay away from outright poison and - above all, and especially if you're a male - don't be fat.