摘要：As scientists working and publishing in the field of front-of-pack nutrition labelling (FOPNL) for many years, we have read with interest and concern the narrative review regarding their effectiveness by Muzzioli et al. [
1].
First, the authors appear to consider that the premises underlying FOPNL—to inform consumers on the nutritional composition of foods from a health perspective and orient them towards healthier purchases—are not the object of consensus in the scientific community, despite them being clearly stated by the WHO [
2]. In addition, they appear to place purely informative labels (such as the NutrInform Battery) and interpretive labels (Nutri-Score, Warning labels, Multiple Traffic Light) at the same level, while there is a clear scientific consensus that interpretive front-of-pack labels are more effective and equitable, and should be promoted (WHO principle n°7) [
2]. Finally, FOPNL have been shown to act as drivers for reformulation, improving the overall nutritional environment [
3].
Second, while FOPNL research usually relies on a clear theoretical framework [
4], and grades the evidence provided by each type of study, the review conflates results from experimental studies, consumer surveys, cohort studies, and even studies not conducted with consumers at all to present a perspective that lacks a clear objective.
Finally, multiple studies investigating the performance of labels, including a very important network meta-analysis [
5] and studies conducted in multiple countries comparing the effects of various types of labels, were entirely omitted from the review [
6,
7,
8,
9,
10,
11,
12,
13,
14,
15,
16,
17,
18,
19,
20,
21,
22]. These studies provide outstanding information as to the comparative effects of FOPNLs to help consumers identify healthier foods and make healthier purchases, and contribute to reduce the burden of nutrition-related disease [
23,
24]. Of great concern is that when studies on performance are referenced, they are misrepresented, instead reporting the results of preference elements within those same studies. For example, the study by Hagmann et al., which found a clear higher performance of Nutri-Score over all other labels, is relayed as finding that ‘most of the participants […] considered the Nutri-Score the least useful’ [
25], which clearly misrepresents the results of the study.
These elements lead to an overall biased and misleading view of the literature, diminishing the value of multiple studies consistently showing that interpretive labels (Nutri-Score and Warning labels, in particular) lead to healthier food choices, as well as overstating the merits of non-interpretive labels, which have consistently been shown to be unable to produce significant modifications in dietary choices.
That FOPNL are only one of the many policies necessary to tackle the obesity epidemic has never been in doubt. However, it is also clear that FOPNLs are a useful health policy tool and, more importantly,
interpretive labels such as the Nutri-Score constitute an evidence-based health policy tool [
26], as reflected in the multiple studies that were apparently overlooked or misrepresented in this study.
