Conflict of interest: The author received no fee nor grant for this commentary. He is personally omnivorous and practices thoracic and cardiovascular surgery.
There is, obviously, since the end of WWII, a battle in the field of human nutrition: animal products are vilified by some academics on the grounds of studies which merely are observational.
The same academics took power and heavily influenced the nutritional guidelines. This is why official recommendations regarding of human health and political decisions stir up some very disputed assertions. This short opinion published in the BMJ[1] sums up them very interestingly. Tim Spector and Christopher Gardner (subsequently named the “authors”) discussed the recent recommendations issued by Canadian researchers from Mac Master University[2] that there is no evidence to cut on red meat because of health risks. They warn against the work of these Canadian researchers who recently demonstrated that one might be skeptical on scientific evidence of the red meat bashing. It is not surprising that such a skeptical attitude is unbearable for those who believe that the weak and inconstant associations between meat consumption and chronic diseases –even death – are non-questionable. It does not matter to them that in those studies the type of meat was very imprecise, as the type of transformation, that those associations were mainly found in quartiles of high consumption and predominate in groups of people with risky lifestyle (male sex, smoking, alcohol). More generally this debate, whatever its polarization, is not only a methodological one or a civilization issue; it is simply a question of efficiency in the actual battle against diabesity. As a physician, this is the aim of our commitment to patients because diabesity is a public health priority. The worldwide prevalence of overweight and obesity has approximately doubled since 1980. Nearly a third of the world’s population is now overweight or obese. Obesity rates have increased in all ages and both sexes irrespective of geographical locality, ethnicity or socioeconomic status. This trend is similar across regions and countries, although absolute prevalence rates of overweight and obesity varied widely. In this battle, this is not an opinion, academic authors and governments failed to tackle the problem by promoting their well-known food pyramids[3]. On the other hand, red meat consumption is not at the origin of this number one problem in public health.
It is simple but, for scientists, it is fundamental to point out that the authors began their piece with significant confusion.
Bacon is not red meat. In mixing them authors perfectly exemplified the huge risk of error in observational studies. The reason why those studies are imprecise, weak and without any evidence of causality whatsoever. On the contrary, randomization will balance this kind of mistakes and control of diet will suppress them. In mixing bacon, red meat, processed meats and environmental issues authors deliberately began to add incomprehensiveness to an already complex issue. As it is their mantra they continued in this path. “The criteria are well suited to studies of pharmaceuticals when there are randomized, double-blind, placebo-controlled trials available, but not for studies of meat, eggs, vegetables or any specific food.” No. It is nonsense to make a rule for drugs and another for diets. Simply this is the heart of the problem. The high-profile authors, evoked in this opinion piece, made their career by making such a double standard an obvious knowledge and at the same time selling their observational studies with a low association between factors as fuzzy as the Mediterranean diet, grains, fat, dairy or meat… It is a shame that observational studies on diet of 10 years or more were published with only one or two assessments by self-reported questionnaires at the beginning of the “prospective” study. Aside from the fact that other academic authors and organisations disagree with this impossibility to do RCT in diets, several very high-quality papers were published and recently. Kevin D. Hall did such RCT which shows that junk food causes weight gain on a short time basis[4]. So, patients and physicians are worth high-quality studies. There are several innovative paths to get rid of observational studies. The first is the RCT. For instance, RCT brings quality information in order to help to solve the obesity epidemics that high profile nutritionists failed to curb. It is untrue to assert: “Given that RCTs with foods or food groups are not possible”. By designing RCT it is possible to precisely assess each type of food or food group in the setting of a clear hypothesis. The second is the use of genetic variants as instruments to test the causal associations[5]. The third is personalised nutrition which is the way to discriminate the effects of nutriments or carcinogens on different genomes and epigenomes[6].
Authors chose the weakest evidence, aka association of factors because they are reluctant to embark on causality studies.
Furthermore the authors made two assertions in the same sentence which are not relevant to red meat.
“The authors also chose to leave out all artificial ex-vivo lab studies and animal studies, included in other meta-analyses, that have, for example, implicated nitrites or acrylamides as carcinogens. “ The carcinogens quoted are totally out of the subject. Is it necessary to point out that nitrites and acrylamide are not present in red meat? How this could be possible in a peer-reviewed opinion? Technically, processed meat means pork, beef or duck that has been salted and cured, with or without smoking. A fresh pound of red meat is not processed. A hard stick of cured salami is. The health risk of bacon, if any, is largely to do with two food additives: potassium nitrate (also known as saltpetre) and sodium nitrite. On the other hand, major food sources of acrylamide are French fries and potato chips; but also, crackers, bread, and cookies; breakfast cereals those last foods that high-profile nutritionists advise eating if they are made of “whole flour”. Furthermore, animal studies and other “mechanistic studies” are not relevant to the new strict analysis of the observational studies about red meat by Canadian authors. Indeed, animal and mechanistic studies are only warning studies for launching clinical trials. No more no less. And another time the great carnivores do not exhibit cancers or CVD after eating meat all their lives. An example of flaw animal study is to give meat to a non-carnivore as it is a poor mechanistic study to give a high-fat diet to rabbits to explain atheroma. Last but not least they didn’t mention the 2010 review which showed that eating charred, smoked, well done red meat is associated with a higher risk of colorectal cancer only in smokers[7]! Interestingly one of the studies reviewed in this paper[8] didn’t find any association between well done or very well-done meat (processed or not) and health issues below 5 steaks/week. We are very near to the advice of the Canadian authors who concluded that until 3 steaks per week there is no measurable risk. In order to understand the complexity of the issue of cancer and meat consumption we need to be clear about the potential causes. It is rather surprising that they didn’t mention the proven carcinogens in high temperature cooked red or processed meat: heterocyclic amines and polycyclic aromatic hydrocarbons. These organic compounds which are also produced by high-temperature cooking of white meat, fish and vegetable protein although in less quantity are not present in fresh meat nor in low temperature cooked meat.
Then they came to the demonstration that Canadian authors cherry-picked studies in their meta-analysis.
Authors argued that Canadian researchers eliminated some observational studies in order to bias the final results. They point out the eviction of Lyon heart study and PREDIMED but they don’t prove that taking into account these studies would have changed the conclusions of the revision meta-analysis about red meat. Doing that, they again delegitimated observational studies. Indeed, meta-analysis of RCT are far less subjected to contest; if red meat consumption is the hypothesis to test, you cannot eliminate an RCT which stated that. In observational studies, the selection is a very complex process as their design and the modelling of multivariate analysis influence the results. Eventually, they agreed with Canadian authors: “Previous summaries, which have been more inclusive, have shown modest health risks for red meat, but clear links between processed meat, heart disease and early death. The evidence for cancer causation is less clear-cut, and several trials of meat reduction have not resulted in lowering of cancer risks. “ In the same paragraph, they also recognize that the issue is not red meat but junk food, a point we strongly agree with. Nonetheless, RCT are very rare despite being more powerful than dozens of observational studies.
Then come the following statements which are only authors’ opinion.
They wrote: “but in our view, we should all be eating less meat and more high fibre, nutrient-dense foods that include many vegetables, legumes, fruits and nuts and seeds.” We were waiting for a better backing than wishful thinking and political correctness. This advice is no more than a copy/paste of the average, standardized, non-personalised nutritional guidelines which proved to be meaningless and inefficient. Nonsense.
Back to the opinion piece I want to point out two errors that were made at the end.
First, why reignite a controversy in a biased manner?
They compare the review articles about meat with an, according to them, a complacent subsidised paper about sugar from the same lead author. Making it short, the Canadian meat meta-analysis would be corrupted by a precedent study on sugar because of hidden conflict of interests. It is a concern. But one may wonder why the EAT-Lancet initiative that they also quoted is not under the same scrutiny? It is suspicious to imply that government subsidies, non-profit organisation subsidies or any other subsidy are neutral… In order to get out of this biased debate, I think we don’t need such a suspicion. It is only a matter of good sense or bayesian evidence: we eat sugar approximately since 1800 and meat since at least 500 000 years. That could help in finding the truth.
Second, it is hard to believe that the authors decided to conclude their opinion piece by a so predictable greenwashing.
The EAT-Lancet initiative that they quoted, is not a new study. It is not a meta-analysis. It is indeed a very sophisticated but biased piece of statistical manipulation in order to impose not less meat but a vegan diet. Their opinion is respectable as long as they state it sincerely. I think that this kind of reasoning is reductionistic as it leads to blaming all people around the planet for eating too much meat. While for instance, south European people eat a very moderate amount of meat and numerous other populations don’t eat sufficient animal products and suffer several deficiencies. It is the same reasoning which is used by bad teachers who punish all the classroom instead of targeting those who are making the heckling. Again, this assertion is not backed by evidence. All agricultural practices have been found to have a variety of effects on the environment. Some of the environmental effects that have been associated with meat production are pollution through fossil fuel usage, animal methane, effluent waste, and water and land consumption. The issue is destructive farming, not whether you eat meat or dairy or vegetables or grains. Another issue is the quantum of those impacts. Contrary to different assertions in the different EAT-Lancet papers, the matter of anthropogenic emissions is extremely complex and still poorly known in details. Particularly meat production is a multifactorial process which is managed very differently in different countries and according to ancestral or industrial practices. The amount of direct GHG emissions of animal husbandry is 2,3 GigaTons of CO2 equivalent id est 5% and indirect GHG emissions linked to all meat, eggs and dairy products are 4,8 gigatons of CO2 equivalent per year or 9.5% of all anthropogenic emissions. It is important to address these emissions with rationality that is why indirect emissions are the true target, not animals. Indirect emissions are the most important and are largely improvable through energy transition and other agricultural practices, the same means used in other sectors of agriculture. Direct emissions are less avoidable but represent only half of the indirect GHG emissions.
As a matter of fact, in the United States 60% of calories come from refined sugar, refined grains and refined omega-6 seed oils, which is saying that meat consumption does not have a prominent role in obesity and type 2 diabetes. It is remarkable that carbs and vegetable oils are recommended by food pyramids. In a recent paper[9], C. Christie et al brought the evidence that excess calories, whatever it comes from, do have a big environmental impact. Among those excess calories, processed foods have a bigger impact in general because they need more energy to be produced, packaged, transported and finally more frequently treated as waste. Excess calories are consumed in developed countries[10], but it is not meat which brings calories in excess because meat and poultry are composed of naturally occurring water, muscle, connective tissue, fat, and bone. The muscle is approximately 75% water (although different cuts may have more or less water) and 20% protein, with the remaining 5% representing a combination of fat, carbohydrates, and minerals. On the contrary, sugar and fats in processed foods, which are in excess, are packed in very dense foods with minimal water increasing significantly the number of calories per day.
Finally, we have to recall the nutritional advantages of meat for humans a lifelong.
Meat is a very good source of nutritionally complete and digestible proteins, bioavailable iron, zinc, and selenium and the exclusive source of vitamin B12. Consumption of red meat has decreased since 2000 in developed countries and the average consumption in Europe is around three portions a week. This amount of red meat is not associated with any increased risk in observational studies. When consumed, meat and offal have an important role in maintaining good health through its supply of high-quality nutrients. In the setting of a flexitarian diet like shown in the blue zone countries, the benefits of meat add those of fish and seafood, more high fibre, plant-based foods, with the nutritional durability of an omnivorous diet. Red meat bashing is against human ancestral diet and will increase human health issues. We should think twice before throwing away red meat production together with fossil fuels because of environmental issues. There are smarter paths to curb anthropogenic emissions without endangering human health.
[1] https://blogs.bmj.com/bmj/2019/10/09/bacon-rashers-statistics-and-controversy/?fbclid=IwAR2453O8TZURygfDz3cz4871bsXTAHhHJ16mhKDHnq7p_3-zjJacTRMmnII
[2] https://annals.org/aim/fullarticle/2752328/unprocessed-red-meat-processed-meat-consumption-dietary-guideline-recommendations-from
[3] https://www.ncbi.nlm.nih.gov/pubmed/30253139
[4] https://www.ncbi.nlm.nih.gov/pubmed/31105044
[5] https://diabetes.diabetesjournals.org/content/early/2019/06/05/db19-0153
[6] https://www.bmj.com/content/361/bmj.k2173
[7] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2769029/
[8] https://www.sciencedirect.com/science/article/pii/S0027510702001641?via%3Dihub
[9] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5799208/
[10] http://www.fao.org/faostat/en/#data/FBS/visualize
This post is also available in: FR (FR)