How that may work: The combination of dietary fat and high-glycemic-index carbs throws a one-two punch. Fat ratchets up insulin resistance making it more difficult to clear blood glucose resulting from easily digested starch. The pancreas releases more insulin to compensate, leaving some with undesirably high levels of insulin and other growth (including cancer growth) promoters.
Fast forward to 2008:
Dietary Patterns As Identified By Factor Analysis And Colorectal Cancer Among Middle-Aged Americans, American Journal of Clinical Nutrition, 2008
"We observed that both for men and (especially) women, a dietary pattern characterized by frequent meat and potato consumption was associated with an increased risk of colorectal cancer."A decade later and this association hasn't gone away. (Although the gender risk has flipped. Sex hormones meddle in strange and wonderful ways. Well, at least when they're endogenous. Synthetic sex hormones like bisphenol-A that leach from bottles and plastic linings of canned foods and which disrupt endocrine functioning are not wonderful. I wonder if this is contributing to gender risk differences. How can it not?)
I like this study. It's one of a new type that analyzes dietary patterns instead of single nutrients. Since we don't eat foods in isolation, drawing meaningful associations from consumption of a single nutrient can be difficult. (Pattern analysis isn't exactly straight-forward either, from what I read, but it's developing.)
In this study, patterns weren't created and overlaid over people's diets, as in, "Who ate the most meat and who ate the most potatoes?" Patterns were instead detected, as in, "People who ate the most meat also ate the most potatoes."
Their analysis identified three major patterns in this group of over 492,000 AARP members (a very unique group):
- Meat and potatoes
- Fruit and vegetables
- Diet foods
I didn't see a rice and beans pattern. Class, culture, and how they impact dietary choices must affect this type of analysis. It would have been nice to see what a rice&beans pattern did to colon cancer risk.
Two more studies. These are meta-analyses, i.e. studies of studies.2
1. Meat Consumption And Colorectal Cancer Risk: Dose-Response Meta-Analysis Of Epidemiological Studies, International Journal of Cancer, 2002
"High intake of red meat, and particularly of processed meat, was associated with a moderate but significant increase in colorectal cancer risk."2. Systematic Review of the Prospective Cohort Studies On Meat Consumption And Colorectal Cancer Risk: A Meta-Analytical Approach, Cancer Epidemiology, Biomarkers and Prevention, 2001
"A daily increase of 100 g of all meat or red meat is associated with a significant 12–17% increased risk of colorectal cancer."The increased risks were nowhere near what the Seventh-day study found, but...
"A significant 49% increased risk was found for a daily increase of 25 g of processed meat."
1. Their comparisons were made primarily among meat eaters, low-consumers vs. high consumers, where some Seventh-day ate no meat at all. The difference between nothing and something may be bigger than the difference between a little and a little more.
2. Also, returning to my original premise, older studies were probably not investigating the pattern of eating. If it's true that fiber (soluble, insoluble, resistant starch each with their own effect) protects the colon lining, if it's true that decreased transit time protects the colon lining (via fiber, fluid, fitness), if it's true that antioxidants and other chemicals in foods (chlorophyll in spinach, resveratrol in red wine) protect the colon lining, if it's true that the combination of fat, especially saturated fat, and high-glycemic carbs is harmful to the colon, then it's going to be difficult to point a finger at any single dietary input.
It doesn't mean dietary risks don't exist, or that we can never discern them. It's as shaun said, "It's like a spider web, when one variable changes, everything else shifts to varying degrees. It's complicated."
2 Meta-analyses are good in that they combine (and so have a large data base of participants) and reassess similar research questions (Does X impact Y?) They're risky in that they only analyze published studies, which creates bias (based on publication bias). Bias can also be introduced in the selection of studies, and in subjective differential weighting of variables. It's always a good idea to check the affiliation of the authors.