Discussions about the role of meat in the diet are very much in vogue. The WHO recently classified processed meat as a class 1 carcinogen (ie – cancer-causing) and unprocessed red meat as a probable carcinogen. Questions of equity, environmental impacts, and animal welfare related to the consumption of meat have also meant that people are shying away from its consumption.
A recent paper appears to fly in the face of this, due to its recommendations that individuals should continue their consumption of red and processed meat. There’s been some (predictable) backlash for various reasons, and some (similarly predictable) meat-eating triumphalism. Some of the criticisms of the paper question its methods of analysis, the data included, and the interpretation of the results (see here). I’ll address a little bit about that as I go, but I don’t think that all of the brouhaha is entirely warranted, even if some of the criticisms may hold some weight.
The first thing I’d recommend anybody truly curious about the topic to do would be to read the paper at hand. It’s open-access – you can read the full text here. Much of the (mis)interpretation of the paper can be put to bed simply by reading it. For those of you that won’t do that, or need some help joining the dots, though, here’s a brief rundown of what they did.
The researchers conducted systematic reviews on the relationships between unprocessed and processed red meats and cardiometabolic health and cancer risk (a total of 4 reviews). They also conducted a systematic review of literature examining the attitudes, values and preferences of people relating to red/processed meats, bringing the total number to 5.
Randomised trials and cohort studies with >1000 participants that ran for >6 months, where differing quantities of red/processed meats were consumed were included in the research, with papers on people already with cancer/chronic disease excluded. The inclusion criteria for randomised controlled trials have been criticised by some as being too strict, and causing the unnecessary exclusion of possibly relevant data.
The estimations of risk reduction were framed around a “realistic” reduction in red meat consumption. Average red meat consumption in North America is 2-4 servings/wk, and a realistic reduction was deemed to be reducing weekly consumption by 3 servings.
A criticism of prior recommendations around meat consumption is that they were overly reliant on observational research, and therefore were limited for drawing causal inferences about red meat consumption. Whilst observational research is a reality of the field in nutrition (it is hard to blind participants, or randomise them to eat a certain diet for long enough periods for differences in chronic disease risk to manifest, and may be unethical. See this article for some more about that), the imprecision of measuring intake and potential for confounding to persist even after correction makes such evidence typically weak.
The certainty of evidence was rated in the current paper using a tool (GRADE) that rates conclusions from “very low certainty” (where the true effect is probably markedly different from the observed effect) through “high certainty” (the authors have a high degree of confidence that the observed and true effect are similar). Whilst studies give an estimation of the strength of relationships between two variables, in matters such as chronic disease that are multifactorial, and where not all relevant determinants can be measured precisely, it is possible (perhaps even probable) that the observed effects within the confines of the study don’t represent reality.
One of the big criticisms of this paper is the use of the GRADE tool, because it rates findings from randomised clinical trials as higher quality evidence, and those from observational/epidemiological research as lower quality.
Whilst I consider that reasonable on-face, nearly all long-term nutritional research is observational (again, discussed here). It’s not feasible or necessarily ethical to randomise people to diets for the very long term, let alone whilst blinding them +/- researchers to what they are eating. Because the outcomes that we tend to be most interested in (incidence/mortality of cancers, CVD, diabetes etc) take years or decades to manifest, usually, research with actual clinical endpoints will nearly always be observational, making it hard to draw causal conclusions (also discussed later). In nutrition research a very large bulk of studies are observational, and penalising well-conducted observational research by automatically ranking RCTs (however well or poorly conducted) as better quality has been said by some to be unreasonable, although that view is not universally held.
This particular analysis compounds the problem, if it is one, because when considering the evidence the “critically important” outcomes for developing recommendations were “all-cause mortality, major cardiometabolic outcomes (cardiovascular mortality, stroke, myocardial infarction, and diabetes), cancer incidence and mortality (gastrointestinal, prostate, and gynecological cancer)” as well as quality of life/willingness to change. That is to say, the outcomes most important for forming recommendations were likely to be near-exclusively observed amongst the research that would be rated as lower-quality. The outcomes that you might reasonably be expected to see differences in over the shorter course of an RCT (surrogate outcomes such as bodyweight, lipids, blood pressure) were rated as “important”. That is to say, the higher-quality evidence would largely speak to less-important outcomes.
Having now given voice to that criticism, I don’t think that those decisions by the research team are unreasonable, I think it just speaks to the nature of nutritional research. That the strength and nature of relationships between individual foods and clinical outcomes is not always well-illuminated by the research available to us is just a fact. Acknowledging that nutrition research is very difficult to conduct, and being mindful of the perils of drawing strong conclusions from data rife with limitations seems better to me than simply holding nutrition research to a different, lower standard than other fields.
All of that said, a 14 person panel was formed, including a research leadership team, health/nutrition/dietetics researchers, and 3 members from outside of the medical and health-care community. The recommendations of the panel were written for individuals, not for a policy-level discussion. This is again of some importance. Relatively minor-seeming reductions in risk at the population level make a large difference, but for an individual may seem trivial. If 12 in every 10,000 people suffer an illness, for a population of 10 million people, reducing the incidence of disease by 2 per 10,000 means 2000 fewer cases per year. However, there’s no way for an individual to know whether they’ll even be one of those unlucky 10 to 12 people. Because the recommendations of the paper do not speak to policy per-se, they also do not preclude policies designed to reduce meat consumption in favour of other foods.
Unprocessed red meat – The certainty of evidence for all measures was low to very low. Diets lower in unprocessed red meat may have little or no effect on the risk of major cardiometabolic disease or cancer. Dose-response studies suggest very small reductions in risk for major cardiovascular diseases and diabetes with decreased red meat consumption. Similarly, very small reductions in cancer risk were observed with decreased red meat consumption.
Processed red meat – Low to very low certainty evidence was found for very small risk reductions in all-cause mortality, cardiovascular disease mortality, stroke, heart attack and type 2 diabetes with a reduction in processed red meat consumption. Similarly, low to very low certainty evidence was found for a reduction in overall lifetime cancer mortality and incidence of a few cancers, and no difference for a number of others, when reducing processed meat consumption.
Attitudes to meat consumption – Omnivores enjoy eating meat (this is unsurprising) and most consider it an essential part of a healthy diet. Many feel that they lack the culinary skills to cook palatable meals without meat.
Assessing causality – As a way of assessing the plausibility of a causal relationship between meat consumption and cardiovascular disease/cancer, the strength of associations observed between meat consumption and clinical outcomes were compared between studies where meat intake itself was measured, and those in which dietary patterns were measured. The logic being that if meat itself is the causal factor, studies examining it directly will see stronger relationships. The opposite was observed to be true, with dietary pattern studies reporting stronger associations.
The exact recommendations of the paper were as follows
Recommendation for Unprocessed Red Meat
For adults 18 years of age or older, we suggest continuing current unprocessed red meat consumption (weak recommendation, low-certainty evidence). Eleven of 14 panelists voted for continuation of current unprocessed red meat consumption, whereas 3 voted for a weak recommendation to reduce red meat consumption.
Recommendation for Processed Meat
For adults 18 years of age or older, we suggest continuing current processed meat consumption (weak recommendation, low-certainty evidence). Again, 11 of 14 panel members voted for a continuation of current processed meat consumption, and 3 voted for a weak recommendation to reduce processed meat consumption.
The reasoning behind the recommendations to continue consumption of red and processed meats were as follows. The certainty of evidence for adverse health outcomes with consuming red meat is low, and it is questionable whether meat has a causal role in the development of CVD and cancer. Presuming a realistic reduction of meat consumption, the absolute reduction in risk is likely to be very small. Finally, because meat-eaters typically enjoy it and consider it an essential part of the diet, trivial risk reductions are not likely to motivate them sufficiently to cut out meat. Finally, the recommendations were framed as “weak”, as the observed attitudes regarding meat consumption were highly variable, meaning for certain individuals reducing meat consumption may still be plausible. A minority (3/14) panellists voted for a weak recommendation to reduce meat consumption.
Whilst the panel stated that their recommendations differ from other guidelines, “[they don’t] seriously challenge them”.
So what does it all mean
Well, it’s all pretty clear-seeming. Red and processed meats don’t appear to pose a massive health risk, and you should keep eating them, right?
I don’t think that that’s entirely the case. Given that no risk estimation actually FAVOURS meat consumption, I wouldn’t consider this sufficient evidence to promote it. It may still be pragmatic to reduce meat consumption, although I certainly wouldn’t go calling meat itself the bad guy.
A weakness of observational research in nutrition is that diets high in one thing by definition need to be low in others. Diets where meat predominate tend to be low in fruit, vegetable and wholegrain matter. Whilst this muddies the waters of interpreting those studies (are the observed risks to do with the presence of one thing, or the absence of another?), it also gives a useful marker for practical diet planning. That research examining dietary patterns high in meat consumption found stronger associations with negative health outcomes than studies examining meat itself only reinforces the point. I’m all for the consumption of meat, I personally love it, but this shouldn’t come at the absence of what we could reasonably believe to be health-promoting components of the diet, being vegetable matter and wholegrains.
If you currently eat a diet that is higher than average in meat content, moderating it via the inclusion of more vegetables (or adding vegetables in in lieu of “less healthful”/less nutrient-dense foods) is still probably good practice. If you only eat a little bit of meat, but not much vegetable matter, you’ll probably get more bang for your buck adding the vegetables in than cutting the meat out.
Similarly, diets that lack vegetable matter are often poor for weight control. Read this article in which I speak a little bit about why my preference for weight management is to favour consumption of vegetables and wholegrains (as well as lean meats and dairy, it must be said). Diets that are high in animal protein, highly processed foods (which have been recently demonstrated to promote greater energy intake) and fat are likely to promote risk of ill-health simply by their effects on bodyweight, given that obesity is one of the greatest determinants of cardiovascular disease, type 2 diabetes, and cancer risk. Whilst observational research that associates high levels of meat consumption with adverse health outcomes might make errors of attribution, it does demonstrate that overall dietary quality is important, and so my general recommendation remains to make sure to get plenty of vegetables and wholegrains in for health, whether you choose to eat meat or not.
This review didn’t look at ethical and environmental factors that might influence decisions to eat meat, although the authors did acknowledge that these are important. The weak recommendation to continue meat consumption, read in context, should not be misinterpreted as saying that these concerns are invalid. The attitudes and values surrounding meat consumption are broad and the recommendations pertain to its health effects. The paper is aimed at individuals making health decisions, not policy makers, and so it may be that for such reasons the production and consumption of meat is reduced on a larger scale without necessarily contradicting the information that the paper is intended to provide individuals.
Finally, the researchers were not able to differentiate between cooking methods of meat and their health effects. It has been proposed that high heat cooking, especially over open flames, generates carcinogens in meat. Those concerned with their health that do choose to consume meat may be advised to use alternative cooking methods.
Eating some meat is probably fine. If there are independent increases in risk of cancer or cardiometabolic disease from meat consumption, they are likely small. The evidence in this paper does not suggest promoting the consumption of meat, and fully-informed individuals may consider a small possible reduction in risk sufficient motivation to reduce meat consumption. Overall dietary balance, including sufficient vegetable matter, still remains advisable. You may still for ethical or environmental reasons choose to limit meat intake, also.
This article has been on my desktop in draft form for a couple of weeks now. During that time, the paper at hand has been discussed by one of my favourite podcasts, Stronger By Science. You can check out the episode here.
They also mentioned this article on examine.com, which includes an excellent breakdown of the study at hand (including the individual analyses on the consumption of meats) and some further discussions about the difficulties and nature of nutritional research.
If you LIKED this article, join my mailing list – it’s FREE and I can send you stuff directly to your inbox. You can also follow me on Instagram, where I frequently post training and diet analysis and advice on my stories.
I also have a podcast, Weakly Weights, available on iTunes and Podbean, where I discuss training for powerlifting in-depth. You can join our mailing list, too, for free sample programs when we discuss them.
Finally, if you like this content and want to learn more from me, check out Fitness Fundamentals – a website providing the most up to date, applicable fitness information, run by my colleague Luke Tulloch and with content written by me.