Relationship between front-of-pack labeling and nutritional characteristics of food products: An attempt of an analytical approach

Machine Learning Approach to Comparing Fatty Acid Profiles of Common Food Products Sold on Romanian Market

Food composition issues represent an increasing concern nowadays, in the context of diverse food commodity varieties. The contents and types of fatty acids are a constant preoccupation among consumers because of their reflections of nutrition and health problems. This study aims to find the best tool for the rapid and reliable identification of similarities and differences among several food items from a fatty acid profile perspective. An acknowledged GC-FID method was considered, while, for a better interpretation of the analytical results, machine learning algorithms were used. It was possible to develop a recognition model able to simultaneously differentiate, with an accuracy of 79.3%, nine product types using the bagged tree ensemble model. The low number of samples or some similarities among the classes could be responsible for the wrong assignments that occurred, especially in the biscuit, wafer and instant soup classes. Better accuracies values of 95, 86.1, and 97.8% were obtained when the products were grouped into three categories: (1) sunflower oil, mayonnaise, margarine, and cream cheese; (2) biscuits, cookies, margarine, and wafers; and (3) sunflower oil, chips, and instant soup.

A Little Bird Told Me… Nutri-Score Panoramas from a Flight over Europe, Connecting Science and Society

Within the Farm to Fork Strategy, the European Commission ask for a unified Front Of Pack nutritional label for food to be used at the European level. The scientific debate identified the Nutri-Score (NS) as the most promising candidate, but within the political discussion, some Member States brought to attention several issues related to its introduction. This misalignment led to a postponement of the final decision. With the aim to shed some light on the current stances and contribute to the forthcoming debate, the objective of the present work is to understand to what extent scientific research addresses the issues raised by the general public. We applied a structural topic model to tweets from four European countries (France, Germany, Italy, Spain) and to abstracts of scientific papers, all dealing with the NS topic. Different aspects of the NS debate are discussed in different countries, but scientific research, while addressing some of them (e.g., the comparison between NS and other labels), disregards others (e.g., relations between NS and traditional products). It is advisable, therefore, to widen the scope of NS research to properly address the concerns of European society and to provide policymakers with robust evidence to support their decisions.

Front-of-pack labels: "Directive" versus "informative" approaches

Front-of-pack labels (FOPLs) aim at communicating to consumers the health value of food items in support of public health policies. Two main types can be discerned: directive and semidirective FOPLs using color schemes (e.g., Nutri-Score) and informative FOPLs (e.g., Nutrinform Battery). Directive approaches tend to show a "wear-out effect" and, additionally, they tend to have various underlying conceptual problems. Usually, their nutritional scores are calculated using changing, arbitrary algorithms and involve a reductionist set of parameters of debatable validity. Thus, they overstate the effects of selected nutritional factors, such as saturated fat and energy, while overlooking the food matrix and the more holistic aspects of nourishment. Moreover, they do not reflect the portion that is consumed, ignore the preparation steps at home, and fail to serve as a useful basis for composing a healthy diet. Also, so long as the nutritional formulations match the algorithmic standards, they tend to allow ultra-processed products. Thus, this might confuse and mislead consumers. Overconfidence in green-colored labels could even result in unbalanced dietary choices, whereas avoidance of red products may eliminate certain foods from the diet that are rich in essential nutrients (e.g., cheese), leading to opposite results than aimed for. The latter is particularly relevant to vulnerable populations, such as the young, pregnant women, and older adults, or for individuals with specific needs. Taken together, directive FOPLs such as Nutri-Score contradict the declared intent of the European Commission to empower consumers to undertake healthy and balanced diets based on easily accessible and robust information. Although informative systems usually also keep the focus on a few selected nutritional parameters, they have are less paternalizing and obviate the need to classify foods as healthy or unhealthy. They also focus attention on the individual portions that are consumed (even if the definition of portion size remains contentious). Given the importance of dietary patterns, rather than individual foods or nutrients, directiveFOPLs of the Nutri-Score type represent a regretful case of nutritionism. Finally, attempts to associate the adoption of a FOPL with an improvement in the health status are few and mainly applied in virtual settings; none of which are longitudinal, nor have they been able to identify a causal link.