Nutritional labelling: ANSES assesses the feasibility of implementing Rayner's score
At the request of the Minister of social affairs and health, Professor Serge Hercberg submitted a report in January 2014 that recommends France-wide implementation of a nutritional information programme based on the Rayner's score system developed for the UK's Food Standards Agency. This information programme aims to provide consumers with information that will enable them to choose between food products based on their nutritional qualities.
ANSES was asked by the Directorate General for Health (DGS) to assess the technical feasibility of deploying the proposed tool. In the report published today, ANSES concludes that implementation of the nutritional score, as defined by Rayner et al. (2005), appears to be technically feasible but would require supplementary data in addition to the data whose labelling will become mandatory in 2016 under the EU regulatory framework. ANSES also indicates that while the nutritional score seems able to effectively compare a large number of food groups, it does not do so sufficiently for certain other groups (non-alcoholic soft drinks, fats, cheeses, fruit purées, chocolates and chocolate-based products, etc.). Lastly, the Agency emphasises that its work did not aim to evaluate the relevance of using this tool for nutritional labelling, nor its capacity to inform the consumer, to induce changes in behaviour or in the food offer, or its overall relevance with regard to public health.
The National Programme for Nutrition and Health (PNNS) intends to employ specific actions to reduce social health inequalities in the area of nutrition. One of its main goals is to "promote access to foods with high nutritional quality" and to pursue "analyses aiming to more effectively inform consumers with regard to nutrition in order to make it easier for them to make food choices".
Regulation no.1169/2011 regarding consumer information on foodstuffs shall make it mandatory as of 13 December 2016 for nutritional labelling to provide the content of products in calories, fats, saturated fatty acids, carbohydrates, sugar, protein and salt.
In a report entitled "Proposals for a new impetus to the French public health nutrition policy", submitted to the Minister of social affairs and health, Professor Hercberg, Chairman of the PNNS, recommends setting up a nutritional information programme for consumers that could be based on a nutritional score, known as Rayner's score, that rates the nutritional quality of foods.
The Directorate General for Health issued a formal request to ANSES on 23 April for assessment of the technical feasibility of calculating this nutritional score established by Rayner et al. (2005).
ANSES's work
The data used for assessing the feasibility of calculating this nutritional score comes from labelling information collected for Oqali. The goal of this database, managed jointly by ANSES and INRA, is to monitor the nutritional quality of the processed food offer available on the French market over time. To date, almost all processed food categories have already been monitored by Oqali. The Oqali database, which includes nutritional values and lists of product ingredients, enables testing of the feasibility of nutritional score calculation using the data found on packaging.
Based on this, the nutritional score established by Rayner et al., could be calculated for over 12 000 processed product references from 23 different food groups classified into 439 families.
Furthermore, by using nutritional composition data from the Ciqual table, representative of generic foods consumed in France, it was possible to add non-processed products (oil, butter, water, fruits and vegetables, meat, etc.), as well as processed foods not found in the Oqali database, to the nutritional score distribution. In this way, nearly 500 generic foods were studied in addition to the products monitored by Oqali.
Agency conclusions and recommendations
The work conducted and the thresholds presented were estimated based on specifications proposed by the Directorate General for Health, with the sole objective of testing the feasibility and the discriminative power of the score, without taking into account the final use of this type of score in a nutritional labelling context.
According to the results, the implementation of Rayner's score appears to be technically feasible, although certain data needed to calculate the score are not yet found on labels. ANSES also indicates that the nutritional score seems insufficiently discriminatory for certain categories of foods (non-alcoholic soft drinks, fats, cheeses, fruit purées, chocolates and chocolate-based products, etc.).
The Agency also emphasises that its work did not aim to evaluate the relevance of using this tool for nutritional labelling, nor its capacity to inform the consumer, to induce changes in behaviour or in the food offer, or its overall relevance with regard to public health.