**5. Main Limitations of Food Databases: Missing Dimensions for Human Health?**

First, beyond only nutrients, foods are interlinked with cultural identity while playing a key role in many local economies, as highlighted by Dembska et al. [18] in their double pyramids models connecting food culture, health, and climate. These authors and others [18,57–61] call attention to the need of leveraging the various dimensions of foods, which are closely related, under the so-called one-health approach [18,57].

If FDBs are specifically useful for balancing a diet for nutrient composition and fully addressing nutritional needs in human studies, they, however, reflect a reductionist view of foods, viewed as only the sum of nutrients [61], not considering the food matrix effect, and hence, the degree of processing [22]. Therefore, to be a relevant tool regarding human health in the long term, their data should not be used alone, but other parameters should be also considered, such as food form and degree of processing, together with other important food properties.

For example, the newly developed Siga score [59] is hierarchically combined with the first degree of processing, then the food matrix effect, added salt, fat, and/or sugar, and the number of markers of ultra-processing (including some cosmetic additives and non-additive markers) [60]. To be elaborated, this score typically needs not only the food composition data, but also the list of ingredients and the presence or not of added sugar, salt, and/or fat. Such a hierarchical and holistic score should be more considered, because, in the end, it is related to global (environmental and human) health [18,57,60,61]; whereas, food composition only is insufficient to address diets from the global or one-health perspective as needed (e.g., compliance with European Climate Law).
