Comparison of nutrient profiling models for assessing the nutritional quality of foods: a validation study
Bibliographic record
Abstract
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.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.096 | 0.127 |
| Meta-epidemiology (narrow) | 0.003 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.005 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.003 | 0.003 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".