Is meat harmful to health

Meta-analysis analysis

A couple of days ago, a message appeared in the news feeds that the consumption of meat has finally been rehabilitated and no longer threatens human health. Moreover, everyone quoted the recommendations of an international group of experts who stated that you can continue to eat red meat and processed meat products almost without any harm to yourself.

With the question of whether this is really the case and whether there is something strange in the new recommendations, the editorial board of N+1 turned to Anton Barchuk, an epidemiologist, a researcher at the University of Tampere and the Petrov National Medical Research Center for Oncology.

The release of this news was preceded by a whole series of publications of the Nutrirecs scientific group, released in one day by the Annals of Internal Medicine, a major and significant journal in the world of medicine. Among the publications were the recommendations themselves and five systematic reviews. Since these reviews are devoted to the prevention of various diseases and premature death, supplemented with recommendations and addressed to the main readers of the journal – general practitioners, the study had a huge resonance.

It should also be noted that there is a rather impressive composition of authors representing different countries of the world. Among them, in particular, was Gordon H. Guyatt, a Canadian doctor who was the first to use the term "evidence–based medicine" back in 1991.

Despite all this, the recommendations of the Nutrirecs group were met with extreme coldness, and several scientific groups immediately published statements indicating both shortcomings in its work and a complete contradiction of the recommendations to existing data and evidence. Perhaps the most complete criticism is presented on the page of the Harvard T.H. Chan School of Public Health (Harvard T.H. Chan School of Public Health).

Let's try to figure out what is wrong with these publications and why the recommendations of an international scientific group published on the pages of a well-known journal should most likely not be blindly trusted. Let's do this in the form of simple, at first glance, questions and answers.

1. The recommendations are based on several complex systematic reviews and meta–analyses - the pinnacles of evidence-based medicine. How can you not trust them?

The question of how much systematic reviews and meta-analyses should be trusted in an era when tools for conducting this type of research have become available to almost any scientist, regardless of his education and level of training, requires a separate discussion. Let's just focus on the fact that the interpretation of the results of any study is almost always a subjective process, regardless of its quality.

But at the same time, there may be several formal complaints and comments about the quality of the works under discussion.

Thus, in the majority of randomized trials considered in one of the reviews, experiments to limit the meat consumed by participants were not conducted. These were studies of various diets, the endpoints of which did not always correlate with the endpoints of meta-analysis.

As a result, two randomized studies selected for meta-analysis evaluated the Mediterranean diet and a low-fat diet. But only the second study was included in the meta-analysis, and the results of the first one were hidden in the appendix to the publication because it seemed suspiciously positive to the authors.

The quality of the conducted meta-analyses of observational studies is also questionable. So, in one of the works included in the meta-analysis of observational studies, nothing is said at all about the effect of meat on health, but at the same time "fish meat" and the ratio of meat and fish consumption are mentioned. How the authors managed to extract the data for their meta-analysis from this work is anyone's guess.

In some cases, when working on meta-analyses, researchers contacted the authors of the original studies to request additional data from them, but when describing the methods used in the works under discussion, nothing is said about this procedure.

Further, the authors of the papers in the Annals of Internal Medicine used quality assessment criteria, which are more often used for randomized trials. If we apply such an assessment scale for observational studies, then all of them will be of poor quality a priori. If all of them are considered low-quality studies, then it is quite difficult to compare them with each other.

The problem of studying the effect of food on health is related to the fact that randomized trials are impossible in many cases. A person's habit of eating or not eating meat develops over the years and strongly depends on other factors (place of birth, social status, etc.). Therefore, a simple intervention in the form of a recommendation to stop eating meat often changes little in the actual consumption of meat.

In some cases, when comparing groups in the study, the effectiveness of the intervention is low (that is, recommendations not to eat meat) it can be regarded as the absence of the effect of meat consumption on health, even if in fact there is such an effect.

2. But even if the meta-analyses presented are not of the highest quality, isn't the number of analyzed studies enough to draw the right conclusions?

Indeed, many studies are included in the reviews, but the results of meta-analyses published in the Annals of Internal Medicine largely repeat the results of previous studies. If we compare all three new meta-analyses with the previous data, we will see that they differ little from each other.

So, in the new publications, the relative risk of various outcomes – be it cancer, diabetes or death – for people who consume red and processed meat turned out to be higher by 7-20 percent (higher than the probability that a person already has by other indicators – heredity, habitat and others, i.e., in fact, many times less – VM) than for those who abstain from this meat. Very similar results were obtained in the papers under discussion. It is all the more surprising that the authors of meta-analyses came to directly opposite conclusions compared to the conclusions made earlier on the basis of similar data.

3. Well, but the authors of meta-analyses write that the effect of meat on health is so small that it can be ignored. Why is this not a reason to doubt the harm of meat?

It should be noted that the main purpose of epidemiological studies is to identify certain cause–and-effect relationships between phenomena (in our case, between meat consumption and health status) and their quantitative, not subjective assessment. The decision on the significance of the influence of certain factors should still be made by people.

By the way, if we compare the quantitative assessment of the effects of reducing meat consumption for health, they are comparable to screening studies of colorectal cancer, breast cancer, lung cancer and prostate cancer, where the protective effect on cancer mortality ranges from 15-30 percent. At the same time, we must not forget that, unlike the refusal of food consumption, screening entails harm in the form of overdiagnosis, complications of examination and treatment, and also requires much more costs.

But, as a rule, despite all this, both the choice of people and the recommendations support the implementation of most of the listed screening programs.

By the way, the authors of the recommendations decided to add to the list of publications also a review on people's preferences regarding meat consumption. The findings of this study strangely support the recommendations: people who eat meat are so attached to it that they are not ready to give it up even if there is a health risk, so we can recommend them to continue eating meat. If we transfer this logic to smoking, it turns out that those who enjoy tobacco should be encouraged to continue smoking.

4. Well, if the recommendations are wrong, then why then how many famous scientists and doctors from all over the world supported this work?

This is a very good question. One of the several problems of the publications under discussion is an overabundance of experts in the field of creating meta–analyses and recommendations with a shortage of specialists in the field of nutrition and epidemiology of diseases associated with food consumption.

Formalizing the evaluation of research results is very useful, but we should not forget that the essence and subject of research are also extremely important. For example, the groups could be supplemented by scientists who had previously conducted observational and randomized studies on the effects of food on health.

But even among the panel of experts who discussed the recommendations, there was no consensus, and some did not support their published version. It is also important to note that for the expert group, the published recommendations are the first and so far the only work.

By the way, even if, most likely, this did not affect the recommendations, but among the authors of the publication was the director of the scientific institution Texas A&M AgriLife Research, where a whole scientific program dedicated to meat is held.

5. But what's wrong with having a different point of view on the question of meat consumption and its impact on health?

The presence of a different point of view is always welcome in science, including epidemiology. With the help of observational studies, modern epidemiology studies causal relationships between phenomena and systematic errors that can hide these connections or, conversely, show them where there are actually no connections. Often, new research reveals these connections or errors and changes our ideas about what is harmful to health and what is not.

But the problem with the presented recommendations is not in the presence of a different point of view, but in the extremely unsuccessful presentation of a new interpretation of old data. In fact, no new discoveries were made, the results of previous studies and meta-analyses were only confirmed, and it was also concluded that those who eat meat do it for pleasure.

At the same time, conclusions were drawn that directly contradict previous recommendations based on the same data.

The question of other indirect, possible effects of meat consumption on health, which the authors of the recommendations for some reason ignored, deserves a separate discussion.

The decision to publish certain recommendations is a lengthy process that requires not just an analysis of the results of existing work, but also an assessment of the balance of benefits and harms, as well as possible effects from the implementation of these recommendations.

The harm from not eating meat has not been sufficiently studied at the moment, but, apparently, it is minimal in populations of countries where its consumption is extremely high. At the same time, there is also a benefit from refusal, albeit minimal, according to the authors of this study, although quite significant according to other scientists.

Both of these prerequisites should definitely not lead to direct recommendations to eat meat. There are many formulations in which you can discuss the benefits and harms in order to leave a person with a choice.

As noted above, the publication of such radical recommendations in such a popular magazine can have consequences all over the world. But, it seems, keen on preparing such significant publications, the authors did not take into account the most important thing – how their publications can affect people's health.

Portal "Eternal youth" http://vechnayamolodost.ru