| I want to talk about one of my all-time favorite studies where researchers split a bunch of rats into different groups with different diets. One group of rats ate a bacon-based diet with up to 60% of their calories from bacon alone. Another group got a similar diet only using beef, and another ate a diet based on chicken with skin. Two other groups got more processed diets—one based on lard with casein while the other was based on olive oil and casein. All the meats were fried and then freeze-dried before feeding. The rats were also injected with a compound known to trigger colon carcinogenesis. Then, they sat back and watched how the different diets affected how the colon cancer “took”. They made some surprising (to them) observations: - The beef-based diet had no effect on colon cancer progression. Despite being riddled with saturated fat, heme iron, and all sorts of “dangerous” compounds. These rats eating a beef diet did not see accelerated progression of cancer.
- The chicken-based diet did not inhibit colon cancer progression. Despite being lower in saturated fat and heme iron than the beef, chicken with skin failed to inhibit cancer.
- The bacon-based diet was miraculous, actually inhibiting cancer progression compared to all the other diets. It was the most protective. Compared to the lard and casein-fed group of rats, the 30% bacon-fed group of rats enjoyed 10% lower ACF multiplicity (a marker of colon cancer progression, with lower being better) while the rats who ate a 60% bacon diet enjoyed 20% lower ACF multiplicity. This is despite the bacon being high in sodium nitrate and other supposed carcinogens.
Researchers didn’t really have a good answer for why this occurred. Their guess was that the bacon-fed rats had less cancer because the salty bacon made them drink so much water. It’s true that the bacon rats drank more water, but as an explanation for the protective effect seems like a stretch. A few years after the rat study, those same researchers “ran” a study plotting a food frequency questionnaire answers against colon cancer in humans. In other words, they used data from surveys where people were asked to remember what they’d eaten in the last 12 months. Then plotted that against health outcomes in those same survey-answerers. The relationships were flipped with the typical “danger foods” like bacon and beef being associated with poor health outcomes and “health foods” like chicken being associated with better health outcomes. I don’t place much stock in epidemiological studies run by data masseuses. I find controlled trials where the food intake is strictly regulated much more compelling, even if the subjects are small and furry. What do you think, folks? Got any favorite studies or bits of research where mainstream experts were forced to reckon with inconvenient truths? | 
