The consumption of ultra-processed foods is of great concern to consumers and this is reflected in the interest it has aroused among the scientific community. In recent years, dozens of studies have been published linking the consumption of these products to various chronic pathologies, including cardiovascular disease, diabetes, cancer and neurodegenerative diseases. 

The body of scientific research generated so far is sufficiently large to allow meta-analyses to be carried out, i.e. studies of studies with a high statistical component, which attempt to draw unifying conclusions from the evidence generated. However, these meta-analyses are specific to each pathology. The novelty of this study is that it proposes an 'umbrella' review of these meta-analyses, in order to be able to draw conclusions for all pathologies as a whole, so that guidelines and directives can be established with dietary recommendations on the consumption of ultra-processed foods. 

An important point to consider, as is done in this article, is to consider the quality of the original studies and the degree of scientific evidence they can provide. Although an enormous number of studies have been published on the relationship between ultra-processed foods and health, the vast majority of them have low or very low levels of quality (expressed as credibility levels) and grades of evidence (GRADE). This is because they are generally observational epidemiological studies which, due to their characteristics, do not allow cause and effect relationships to be established. In general, studies of this type are considered to have a low level of evidence at the outset. 

It has been suggested that nutrition recommendations should not be made with low levels of evidence and mediocre study quality and that it is necessary to have evidence based on clinical trials and their meta-analyses, which do allow causal relationships to be established. However, as discussed in the article, in the case of ultra-processed foods this is extremely complex for several reasons. One of the main reasons is that it is only possible to conduct such clinical trials for intermediate outcomes such as body weight, insulin resistance, inflammation or gut microbiota, but not for endpoint markers such as the diseases mentioned above, for ethical reasons. It should be noted that the existing evidence is sufficiently robust to suspect that administering such foods to people in a clinical trial could have dire consequences for them. 

Therefore, and this is common in nutrition, for the time being and probably in the future, it will not be possible to have high-level scientific evidence on the consumption of ultra-processed foods. In that sense, the umbrella review shows that, although low grade, the available scientific evidence is very consistent for all studies and meta-analyses published to date. In particular, relative risk levels for various pathologies have been observed to be around 30% and can be as high as 55% for obesity. 

Consequently, the study presented here is important because it reports an increased health risk associated with exposure to ultra-processed foods for virtually all pathologies. The strongest available evidence has been found for associations between increased consumption of ultra-processed foods and all-cause mortality, cardiovascular disease-related mortality, mental disorders, overweight and obesity, and type 2 diabetes. Thus, the authors recommend the implementation of public health strategies aimed at reducing dietary exposure to ultra-processed foods to improve human health.

EN