Meat the Facts: Red Meat Causes Diabetes?

Two major papers were just published back to back claiming a link between red meat intake and type 2 diabetes, and the media has been eating it up like a blooming onion from Outback Steakhouse. Now, as a PhD scientist with a specialty in human metabolism, I've spent quite a lot of time reading the literature on the links between certain foods or food groups and chronic diseases. And I'm here now to explain to you why I'm not gonna give up red meat anytime soon, nor do I think it's a sensible thing to do based on these data to protect myself or yourself from type 2 diabetes. And I'll also tell you why, why I care so much about this topic. And hint, it's not because I like meat. So, let's start with some of the headlines covering these papers, including those from CNN, the New York Post, Healthline, the Guardian, Medical News Today, all have direct and sensational headlines, with some making causal claims like "increase your risk of type 2 diabetes." But the studies had serious limitations. Take the paper published in Nature Metabolism linking heme iron intake, heme iron being in red meat, 2 diabetes. In this study, they looked primarily at 3 cohorts including over 200,000 participants and found a correlation between heme iron intake and type 2 diabetes with a hazards ratio of 1.26, representing about a 26% increased risk of incident type 2 diabetes comparing the lowest to the highest quintiles of heme iron intake. Now that sounds compelling, right? It sounds authoritative. Well, the participants with the greatest heme iron intake in this study were also, as a population, less physically active, had higher BMIs, and were more likely to smoke. The paper even notes that heme iron intake was positively correlated with a Western-style diet, which includes here and now I'm quoting, "more sweets and desserts, French fries, and more refined grains." Now, these represent huge confounders. Of course, the researchers can and did attempt to adjust for these confounders, but adjustments of these sort are assumptive and often incomplete. And when you consider that risk factors like obesity—taking obesity alone constitutes an order of magnitude or more stronger risk factor than the association observed in this study between heme iron intake and type 2 diabetes, I'm already left extremely skeptical. What's more, they found that increased heme iron intake correlates with unfavorable metabolic markers, including higher insulin levels and higher triglyceride to lower HDL ratio. And the reason I find this particularly important to point out is because I think it allows one you to litmus test the relevance of these data on yourself due to the fact that whatever underlying metabolic dysfunction presumably contributes to type 2 diabetes also manifests in these worse markers of metabolic health. So if a person, say, adopted an animal-based ketogenic or even full carnivore diet and their metabolic health markers are good or optimal, or just improving, I'd suggest these data have very little relevance to that person on an individual level. Otherwise stated, if the argument goes that red meat will cause metabolic dysfunction manifesting in insulin resistance, high triglyceride to HDL ratio, and that this metabolic dysfunction increases diabetes risk, well, if one eats a lot of meat and their metabolic markers, suggest they're in great metabolic health, well, you can decide for yourself what you think that means. I think it's pretty clear. In summary, on this first paper, the effect size was small to modest and was dwarfed by the risk imposed by other factors like obesity. The associational data are confounded by obesity, physical inactivity, and smoking, which cannot be assumed to be fully corrected for statistically, nor are many other healthy user bias elements corrected for in this study. That said, I would not fully rule out a connection between heme iron intake and type 2 diabetes. There may be some really interesting pathophysiology here, and I'm not saying that ironically, I'm saying that seriously. But for now, I'd say this is more fodder for hypothesis testing and an academic quirk than clinically actionable information. And I'll hammer home that point after I discuss the next paper, which was perhaps the bigger news paper that was published in one of the Lancet journals. This second paper was a meta-analysis of 31 cohorts and found an association between red meat intake and incident type 2 diabetes with a hazards ratio of 1.10, which is even smaller than the first study. But a few key points: 1. Again, the effect size was small. Tiny. The hazards ratio was 1.10, where 1.0 constitutes no increased risk. And this is compared to something like having a BMI over 30, which has a risk ratio of 4.44. Way bigger than red meat intake. Point 2, and this is really important, there was substantial heterogeneity among the cohorts. In fact, 21 of the 31 cohorts showed no association between unprocessed red meat intake and type 2 diabetes, with some showing actually trends towards lower risk. Now, as an aside, but an important one, on meta-analyses, the immediate counterpoint to this would be, "Oh, but it's a meta-analysis." Otherwise put, one could argue that while there is substantial study in cohort heterogeneity, and the majority of the cohorts had negative findings, the aggregate of the studies is more likely to reveal the truth. However, mixing different studies on distinct populations also sacrifices specificity and has the potential to muddy the metabolic waters. And this is an implicit limitation of meta-analyses. So while it's possible a meta-analysis can help provide clarity, where there are conflicting data, it can also obscure the truth by the process of lumping together different studies with different methods, different confounders, and different populations. So my opinion, simply stated, is metas don't necessarily represent a greater truth. They can also be used to reinforce deception, just with more perceived academic muscle. Okay, aside over. Point 3 on this Lancet study is similar to the other paper we reviewed in Nature Metabolism, this Lancet study has major confounders. The paper itself admits, and this is a quote, "The observed associations could therefore reflect the bias away from the null owing to relatively unhealthy lifestyles associated with meat consumption." That's scientist talk for People who eat more red meat are also likely to live an obesogenic westernized diet, and that can confound the results. Additionally, I personally found attempts at biological plausibility in this paper questionable. They report, "No definitive effects regarding mechanism have been reported," and "The exact mechanism has yet to be established." To me, this reinforces the point. That the mechanistic explanations remain extremely shaky despite an abundance of interest and resources spent on studying this topic, that of red meat and chronic disease. So when you combine all these factors—small effect size, conflicting data across the cohorts, presence of major confounders, and also kind of unclear mechanism—what you're left with is, in my opinion, somewhat of a clinical nothingburger. Now, I did say clinical nothingburger. What I mean by this is the low-hanging fruit for diabetes prevention remains sugar and refined carbohydrate reduction and weight management. And that should not be controversial. In summary, arguments that red meat causes diabetes stand on shaky data at the very best, and they're associational. Two, observed associations may be driven by major confounders. Statistical adjustments cannot be assumed to fully account for these confounders. Three, even if there is an effect, the effect sizes are small and dwarfed by other risk factors like obesity. And finally, the bottom line when it comes to diabetes prevention is that red meat elimination or reduction is not the low-hanging fruit. The point should be reduction of sugar, refined carbohydrates, and weight management. Now to that other question that I promised I'd answer: Why do I care? Because the fact of the matter is, I personally could easily be pescatarian. Yeah, I said it. I genuinely prefer fish over beef, and if you supplied me with infinite sashimi for life, I'd give up ribeye for life without much issue. Basically based on personal taste preference. But my taste preference, whether you agree with them or not, are irrelevant to the data. And I find it really problematic that when studies like these come out, they're sensationalized to drive fear engagement, and to cement status quo narratives around health, including by academics on social media who may title and abstract bomb their followers consistent with their worldview but without doing serious critiques of the data and what the data say or, in this case, do not say. Granted, this is not just an anti-meat problem, although, again, in my opinion, red meat is a popular health scapegoat. The result of this intellectual ecosystem is not just bickering among academics, but clinical harm. For example, we published recently a case series of 10 patients who used carnivore-style ketogenic diets to treat inflammatory bowel disease, ulcerative colitis, and Crohn's disease, and for these patients and many like them, going against the common narrative was actually life-saving. Now, that does not mean by any means that red meat, a red meat heavy diet, is best for all humans. I am not arguing that. But when I reflect on what I read in the literature and observe in social media and see clinically, I typically see more harm than good arise from the sort of shallow thinking that I see reflected in the coverage and responses to these recent papers. So, closing, in my opinion, we not only can do better, but we must do better, because the stakes are too high.