Note: This analysis can be applied to most news reports about the study in question, published in various media outlets, and not only to the news published in Visão magazine, which is presented here as an example: http://visao.sapo.pt/actualidade/sociedade/2019-09-05-Vegetarianos-podem-ter-menos-doencas-cardiacas-mas-maior-risco-de-AVC
1. Facts under analysis?
A recent study, published in the prestigious British Medical Journal (Tong et al., BMJ 2019;366:l4897) , reveals that individuals who follow a vegetarian or vegan dietary pattern have a lower risk of having a myocardial infarction, but an increased risk of having a stroke.
The question is whether the study actually allows us to draw that conclusion, as it has been published in various media outlets.
2. Analysis of the fact(s) based on scientific evidence
Today we are witnessing a phenomenon in Western societies where a growing number of individuals are altering their traditional dietary patterns to a type of diet that is progressively freer of animal products. These individuals, whose diet is designated as vegetarian or vegan, depending on the degree to which they avoid animal products, seek, among other things, to have a healthier diet through this change.
Although this is not the only reason for this change in eating habits, the scientific community has been trying to answer this question, that is, whether and to what extent this type of diet is healthier.
The article in question , published at the beginning of September, seeks to answer these questions, using a cohort of approximately 48,000 individuals residing in the United Kingdom who are part of a larger study, the EPIC study ( European Prospective Investigation into Cancer and Nutrition ), which includes the impressive number of more than half a million individuals from ten European countries, followed for more than fifteen years.
This is therefore a very significant number of individuals, which in itself allows for a robustness in the conclusions that would be difficult to obtain with smaller samples and shorter follow-up periods, as is more common in studies of this kind.
The study's conclusions are clear and point to a lower risk of coronary heart disease in individuals following vegetarian/vegan diets compared to those who consumed meat. This difference remained significant (albeit marginally) when the results were adjusted for confounding variables such as blood pressure, body mass index (BMI), plasma cholesterol, or the presence of diabetes (self-reported variables). This reduction, of approximately 20%, is consistent with previous studies demonstrating the role of LDL (or non-HDL) cholesterol in the risk of myocardial infarction. The present study again observed that individuals who include meat in their dietary pattern have higher LDL cholesterol levels.
On the other hand, a significant increase in the risk of stroke (also of approximately 20%) was observed in individuals with a vegetarian/vegan dietary pattern. This increase was observed mainly for hemorrhagic stroke, with the increased risk of ischemic stroke being of lesser magnitude and without statistical significance. These results regarding stroke were more surprising and less consistent with the literature.
Therefore, the news headline seems factually correct and hardly subject to criticism. However, it's important to understand to what extent we can assert these dietary effects based on a study of this type. In reality, a cohort study like this one does not represent the highest level of evidence available to us and which we need to consistently infer cause-and-effect relationships between phenomena (we can find a good text on the hierarchy of scientific evidence here ). In this case, to have a good degree of certainty that vegetarian dietary patterns reduce coronary heart disease or increase the number of strokes, we would have to resort to randomized, double-blind clinical trials, which, at least in this case, would be almost impossible to conduct. This is both the problem and the fascination that Nutritional Sciences hold , because although epidemiological studies often point us towards certain answers, we almost always lack definitive proof from a clinical trial.
The "essence" that the ingenuity of so many scientists has extracted from cross-sectional or cohort studies has been fundamental to understanding the very complex relationships between diet and health, but this should not lead us to become overly enthusiastic about the conclusions we draw from them. Furthermore, special care must be taken in communicating this information through general media and aimed at the general public, whose capacity to interpret these "nuances" is certainly less.
3. Final verdict of “Thinking Nutrition”
In conclusion, although the news report in question is faithful to the study's conclusions, the degree of certainty (or lack of uncertainty) with which the results are announced is not compatible with the study's intrinsic limitations, nor with the obvious fact that it is only one study. Therefore, the news report is considered "Inaccurate" .
