Diet and Epidemiology: Colon Cancer
Over the past century, as our medical system has grown, I think the influence – both positive and negative – of epidemiology on medicine has also become more pronounced. Epidemiology has always lurked below the surface of medicine – from ancient Indian physicians recognizing that boiled water reduced the incidence of infectious disease, to Dr. John Snow tracing patterns of infection to identify the source of a cholera outbreak in 1854 – and has allowed for sweeping public health interventions that have drastically reduced morbidity and mortality (e.g. smoking cessation). In fact, epidemiology is inseparable from the practice of medicine, as a physician ought to take into account the patient’s race, where he/she lives, what he/she does, etc. because all these factors have been shown to predispose individuals to a variety of diseases.
This model for viewing individuals works well when we are dealing with a single causative agent (e.g. E. Coli from beef); we can sort through patients’ histories and symptoms to find the 1-2 factors (e.g. eating at Jack-In-the-Box) that correlate most strongly with the chance of being infected and, then, treat the patient accordingly, for these 1-2 factors can explain a majority of your risk. After the antibiotic era (which emerged in the 1930s, following the discovery of penicillin) knocked infectious diseases off the list of primary killers (see table below), American began to turn its attention to the diseases that were left: heart disease, cancer, diabetes, et al. (as of 2009, Ref. 1). 
Yet, when we apply epidemiologic principles to these multifactorial diseases, we rarely identify a characteristic of the patient that can explain the majority of their risk. An increased risk for colon cancer, for instance, has been linked to diets high in saturated fat and red meat, with the increased risk estimated to be as high as two-fold. However, it’s difficult to take into account all the other factors that influence cancer risk: the vegetables & fruits you eat; how those veggies are prepared; the amount of exercise you do; what kind of exercise; where you live; how much meat you eat; how your meat is prepared. So, one the first studies (Ref. 2) – now cited nearly a thousand times – that pointed out the dietary association found that chicken-n-fish seems to be protective (decreased risk of colon cancer) compared to red meat-heavy diets. Yet, they only corrected for total energy intake in their model; they plugged vitamin intake, exercise, and fiber into their model one at a time and found that these factors were not influential. With complex phenomena, it’s no surprise that individual factors – especially biochemical modulators like vitamins – have no “significant” effect in isolation. (I would have liked to see what the combinatorial effect of vitamins, fiber, and exercise was.) Jump ahead 20 years and the situation hasn’t improved much: a group (Ref. 3) recently tried to tease apart how cooking methods – rare vs. well-done – affected colon cancer risk, and I’ve shown their Table 2 below. You can see that fruit intake, vegetable consumption, and physical activity all trend inversely with red meat intake. Once again, they found that each of these confounders does not influence colon cancer risk when examined one by one… so, they ignored all of these lifestyle modifiers and told us, again, that red meat increases your risk of colon cancer. (Obviously, grilled meat is more highly correlated)
Is red meat correlated with an increased risk of colon cancer? Sure. But does it cause colon cancer? Hard to say. In these studies which form the backbone for the no-red-meat argument, there are so many confounding variables that it is nearly impossible to attribute the observed effect to any single player. In the end, what these studies indicate is that it’s not red meat alone, but, rather, the myriad factors contributing to a red meat-heavy lifestyle that result in an increased risk for colon cancer. In fact, I think that the dietary link may have more to do with the way food is prepared rather than the food items themselves. For instance, a 6 oz. serving of grass-fed beef paired with an equal (by volume) serving of kale & carrots would (by my estimates) be less harmful than a 6 oz. slab of corn-fed beef paired with an iceberg lettuce salad, a cup of salty soup, a dinner roll, a baked potato, and a few tablespoons of green beans. In light of such dramatic variations in food preparation, we need to think about the cultural & societal norms that lead to such tumorigenic and obesogenic dietary patterns, for this is where the problem lies.
Cutting out red meat from your diet is not the same as not drinking water from a cholera-infected fountain: there is no single player, and simply removing red meat won’t eliminate the risk for colon cancer. Yet, with such few medical breakthroughs nowadays, biomedical research and the popular press leap upon whatever marginal gain they may find. In the end, these kinds of results give the media something to shout about for a time being (e.g. resveratrol, lycopene, ad infinitum) before the next food article demonstrating incremental benefits is “discovered.”
