Comparison of nutrient profiling models for assessing the nutritional quality of foods: a validation study

Nutrient profiling (NP) is a method for evaluating the healthfulness of foods. Although many NP models exist, most have not been validated. This study aimed to examine the content and construct/convergent validity of five models from different regions: Australia/New Zealand (FSANZ), France (Nutri-Score), Canada (HCST), Europe (EURO) and Americas (PAHO). Using data from the 2013 UofT Food Label Information Program (n15342 foods/beverages), construct/convergent validity was assessed by comparing the classifications of foods determined by each model to a previously validated model, which served as the reference (Ofcom). The parameters assessed included associations (Cochran–Armitage trend test), agreement (κ statistic) and discordant classifications (McNemar’s test). Analyses were conducted across all foods and by food category. On the basis of the nutrients/components considered by each model, all models exhibited moderate content validity. Although positive associations were observed between each model and Ofcom (all P trend<0·001), agreement with Ofcom was ‘near perfect’ for FSANZ (κ=0·89) and Nutri-Score (κ=0·83), ‘moderate’ for EURO (κ=0·54) and ‘fair’ for PAHO (κ=0·28) and HCST (κ=0·26). There were discordant classifications with Ofcom for 5·3 % (FSANZ), 8·3 % (Nutri-Score), 22·0 % (EURO), 33·4 % (PAHO) and 37·0 % (HCST) of foods (all P<0·001). Construct/convergent validity was confirmed between FSANZ and Nutri-Score v. Ofcom, and to a lesser extent between EURO v. Ofcom. Numerous incongruencies with Ofcom were identified for HCST and PAHO, which highlights the importance of examining classifications across food categories, the level at which differences between models become apparent. These results may be informative for regulators seeking to adapt and validate existing models for use in country-specific applications.

Townsend, 2010 d Extent to which the measurement correlates with an external criterion of the phenomenon under study at the same point in time. Rayner, 2011 (unpublished manual)  Extent to which the system is a useful tool to the consumer making food-purchase decisions in the marketplace; determined by end users of the system (i.e., the consumer).  FVNL is the third ingredient and is estimated to represent ≥25% of the product's weight f .

>60
FVNL is the first ingredient, but non-FVNL ingredients appear to contribute substantially to the product's weight c .
FVNL is the second ingredient, and the amounts of the first and second ingredients are similar e .

>80
FVNL is the first ingredient, and only FVNL ingredients contribute substantially to the product's weight. FVNL, fruits, vegetables, nuts, and legumes. a In the absence of quantitative declarations in the ingredient list, which are not required for food labelling in Canada, a method was developed by the research group at the University of Toronto in order to estimate the FVNL content and corresponding FVNL points of the foods for the Ofcom model. Given that the ingredients are listed in descending order by weight, the presence and positions of the FVNL ingredients within the ingredient list were used to estimate the FVNL content.
b If sub-ingredients in brackets were presented for an ingredient, the ingredient was considered a FVNL if one of the first two sub-ingredients was a FVNL. If a food consisted of more than one component, and therefore more than one ingredient list (e.g. tuna kit with crackers), FVNL points were calculated for the individual components, and the average of the FVNL points was used to represent the food as a whole. If the ingredient list of a food was missing (<2% of foods), FVNL points were not assigned unless it was evident from the product name that FVNL contributed substantially to the product's weight. c Non-FVNL ingredients that were considered to contribute minimally to the product's weight included: salt, preservatives, colour, vitamins, minerals, oils, flavour extracts, antioxidants, and food additives. Other non-FVNL ingredients were considered to contribute minimally if they appeared after salt or preservatives in the ingredient list. Non-FVNL ingredients that were considered to contribute substantially to the product's weight were other than those previously listed (e.g. sugar and water were considered to contribute substantially to the product's weight). d Concentrated FVNL ingredients are those in concentrated form (i.e. dried, evaporated, as pastes). According to the Ofcom model, concentrated FVNL contribute less than non-concentrated FVNL to the weight of the food; thus, the weight of concentrated FVNL should be multiplied by two when calculating the FVNL content of a food. Given that this criterion cannot be applied directly in the absence of quantitative declarations, the criteria were adjusted for concentrated FVNL ingredients. e A food may score two points if the total number of ingredients present in the ingredient list is low (e.g. ≤3), and if it is estimated that the amount of the first ingredient is only slightly higher than the amount of FVNL in the food (e.g. estimated proportion of 55 versus 45%, respectively). f A food may score one point because at least 25% of the weight represented by a concentrated FVNL equals to at least 40% FVNL when multiplying the amount by two according to the Ofcom model.