PosterPDF Available

The updated algorithm of front-of-pack label Nutri-Score is not in line with Dutch food-based dietary guidelines. Results of calculations with Dutch food composition database

Authors:
  • Dutch Dairy Association
  • Dutch Dairy Association
  • Food Safety & Nutrition Consultancy

Abstract

Introduction In the Netherlands the Front-of-pack (FOPL) Nutri-Score has received a lot of opposition (by approximately 180 food scientists in 2019), because Nutri-Score was not sufficiently in line with Dutch food-based dietary guidelines (FBDGs). In 2022 an updated algorithm has been published with an attempt to bring Nutri-Score more in line with European food-based dietary guidelines. In this study the updated algorithm is compared to the Dutch FBDGs which is called The Wheel of Five. Method The renewed 2022 algorithm for solid foods is applied to the Dutch Food Composition Database (NEVO) to calculate the Nutri-Score values for all foods. The Nutri-Score values were compared to the Dutch food-based dietary guidelines. Foods that are included in the Wheel of Five are considered as “healthy”, i.e. would qualify for labels A or B, and foods that receive labels C/D/E are considered “unhealthy”. Results Figure 1 shows the distribution of all solid foods according to their Nutri-Scores and their place in the Dutch food-based dietary guidelines. In total, 1980 solid foods were selected from NEVO. 19% of unhealthy foods still received a ‘healthy’ Nutri-Score A or B. In addition, 25% of healthy foods scored “unhealthy’ Nutri-Scores C-D-E. Table 1 show examples of foods groups with a large misalignment with the Wheel of Five. These are e.g. fish, cheese, nuts and seeds and cereal products Conclusion In general, the updated 2022 algorithm of Nutri-Score has got a 19% error in evaluating healthy foods as unhealthy and a 25% error in evaluating unhealthy foods as healthy according to their Nutri-Scores for all food groups. Food groups with the largest mismatch are cheese, solid milk products, fish and potatoes and tubers. Controverse After publication of the updated algorithm again more than 200 Dutch food scientists and, in addition, two dietician associations have sent a letter to the Dutch Ministry of Health with a request to not introduce Nutri-Score in the Netherlands and wait for further adjustment of the algorithm until it is sufficiently in line with the Dutch FBDGs to prevent confusion among consumers. Nevertheless, the Ministry has decided to ignore this request and introduce the Nutri-Score as voluntary label as of 2024.
The updated algorithm of front-of-pack label Nutri-Score is not in line with Dutch food-based dietary guidelines.
Results of calculations with Dutch food composition database.
Jacco Gerritsen MSc (1), Dr. Stephan Peters (1), Prof.dr. Hans Verhagen (2)
1. Nederlandse Zuivel Organisatie (NZO)/Dutch Dairy Association, the Netherlands 2. Food Safety & Nutrition Consultancy, the Netherlands
Introduction
In the Netherlands the Front-of-pack (FOPL) Nutri-Score has received a lot of opposition
(by approximately 180 food scientists in 2019), because Nutri-Score was not sufficiently
in line with Dutch food-based dietary guidelines (FBDGs). In 2022 an updated algorithm
has been published with an attempt to bring Nutri-Score more in line with European food-
based dietary guidelines. In this study the updated algorithm is compared to the Dutch
FBDGs which is called The Wheel of Five.
Method
The renewed 2022 algorithm for solid foods is applied to the Dutch Food Composition
Database (NEVO) to calculate the Nutri-Score values for all foods. The Nutri-Score
values were compared to the Dutch food-based dietary guidelines. Foods that are
included in the Wheel of Five are considered as “healthy”, i.e. would qualify for labels A
or B, and foods that receive labels C/D/E are considered “unhealthy”.
Results
Figure 1 shows the distribution of all solid foods according to their Nutri-Scores and their
place in the Dutch food-based dietary guidelines. In total, 1980 solid foods were selected
from NEVO. 19%of unhealthy foods still received a ‘healthy’ Nutri-Score A or B. In
addition, 25%of healthy foods scored “unhealthy’ Nutri-Scores C-D-E.
Table 1 show examples of foods groups with a large misalignment with the Wheel of
Five. These are e.g. fish, cheese, nuts and seeds and cereal products
Conclusion
In general, the updated 2022 algorithm of Nutri-Score has got a19% error in evaluating
healthy foods as unhealthy and a25% error in evaluating unhealthy foods as healthy
according to their Nutri-Scores for all food groups. Food groups with the largest
mismatch are cheese, solid milk products, fish and potatoes and tubers.
Controverse
After publication of the updated algorithm again more than 200 Dutch food scientists
and, in addition, two dietician associations have sent a letter to the Dutch Ministry of
Health with a request to not introduce Nutri-Score in the Netherlands and wait for further
adjustment of the algorithm until it is sufficiently in line with the Dutch FBDGs to prevent
confusion among consumers. Nevertheless, the Ministry has decided to ignore this
request and introduce the Nutri-Score as voluntary label as of 2024.
DSGR45
Food
category Products in the Wheel of Five (%) Products not in the Wheel of Five (%)
Nutri
-Score A B C D E A B C D E
Cheese (n=71)
0 0 27 73 500669 24
Potatoes and tubers (n = 47)
19 70 11 0 0 15 40 45 0 0
Nuts and seeds (n = 37)
18 532 45 000760 33
Fats and oils (n = 70)
0 7 93 000022 41 37
Fish (n = 95)
57 12 615 11 00000
Pasta, rice and other cereal products (n = 14)
86 14 00029 57 014 0
Solid milk products (n=130)
71 18 903528 41 20 6
Figure 1: Percentage of the Nutri-Scores of all product groups* (minus drinks) that are in the Wheel of Five and that are not in
the Wheel of Five, according to the old and new algorithm. (N=1611) * Product groups that fall outside the scope of Nutri-
Score - such as vegetables, fruit, herbs, spices and infant nutrition - are not included
Table 1: Food categorisation according to renewed 2022 Nutri-Score algorithm
Article
Full-text available
The full article is available on request. - Background - The front-of-pack label Nutri-Score is currently proposed as the system of choice in seven EU countries. However, there is still much scientific debate about the validation and efficacy of Nutri-Score and there is much discussion about author affiliation and study outcome. - Methods - Recently we published our paper: Nutri-Score and publication bias: A complete literature review of the substantiation of the effectiveness of the front-of-pack logo Nutri-Score Peters & Verhagen, PharmaNutrition 27 C (2024) 100380. This paper received a commentary paper by the developers of Nutri-Score: M. Touvier et al. 2024 “Rebuttal to the paper published by S. Peters and H. Verhagen”. The rebuttal has also been published on the Nutri-Score blog https://nutriscore.blog/2024/02/19/rebuttal-of-the-claims-against-the-nutri-score-made-by-two-lobbyists-in-pharmanutrition-in-an-effort-to-discredit-academic-research/. We herewith provide an invited commentary to that rebuttal paper, which further supports the observed publication bias. - Results - In this response to the rebuttal, we primarily respond on the scientific issues raised in the rebuttal and explain more about our alleged conflict of interest and our motivation to write the paper. Moreover, we basically thank the authors of the rebuttal paper for, perhaps ironically but essentially, confirming our analysis: there is a publication bias versus affiliation. - Discussion - Overall, the available evidence is clearly limited and biased, and more research is needed to substantiate or disprove the effectiveness of Nutri-Score.
ResearchGate has not been able to resolve any references for this publication.