This is something I've had a great interest in for years and have read many studies where there is clearly correlation but so many obvious confounders that claims of casuality are just not supported by the data. If the summary of the sudy you linked is accurate (http://web.archive.org/web/20090223222003/http://www.nutriti...) it is an example in that they studied people with low fat and high fiber diets, but had no group of high fat and high fiber diet. So is it the fat or the fiber or both? They also made no attempt to distinguish between types of fats. So is it animal fat +/- fiber or any/all fats +/- fiber? Further, they talked about fat and fiber not protein. Ultimately they made no claims about casuality, only correlation. Given more recent reports of the effects on diseases of fiber-fed gut bacteria it might be that it's the fiber that's important.

My personal question is, does a high fiber ketogenic diet (high fiber, high fat, low carb) raise or lower cancer risk? I don't know of strong evidence either way.

Criticism of "The China Study" pretty much ends my confidence in it: https://rawfoodsos.com/2010/07/07/the-china-study-fact-or-fa...

Thanks for that link. It's an excellent analysis of the actual data from the study. I'm part way into reading it and right away the author has thrashed some of the claims of the original study. It appears that even some of the basic claims of correlation aren't even correct.

Added: Schistosomiasis infection is such a huge cofounder for colorectal cancer rates that it is ridiculous to have claimed dietary cholesterol from animal sources as the cause.

That's pretty neat.

Nutrition seems difficult to study because there are so many biological mechanisms at play, but we can get a rough picture at the macro level.

Of course someone always blows findings out of proportion and decides to take things to an extreme.