CO 013. VALIDATION OF THE MEXICAN WARNING LABEL NUTRIENT PROFILE

Background. Different nutrient profiles have been developed in Latin-America to assess nutritional quality of packaged food products. Recently, the Mexican Nutrient Profile (NP) was developed as part of the new warning label regulation.

Objective. To assess the validity of the Mexican NP against the Pan-American Health Organization (PAHO) model using the calibration method in a sample of packaged foods and beverages sold in Mexico. We also compared the classifications produced by these two models against those from Chile, Ecuador, Peru, Uruguay, and Brazil.

Methods. Nutrition content data of 38,872 packaged food products available in the Mexican market were collected in 2016 and 2017. The validity of the Mexican NP was assessed by comparing the proportion of products classified as ‘Healthy’ (without warnings) or ‘Less healthy’ (with one or more warnings) against the PAHO NP. Comparisons between NP from Mexico and the NP of five Latin American countries (NP- Ecuador, Chile (3 phases), Peru (2 phases), Uruguay and Brazil) were estimated through Kappa coefficients, Pearson correlations, and proportion tests.

Results. The three phases of the Mexican NP classified a similar proportion of products as healthy (19.1% to 23.8%) as the PAHO model (19.7%), showing the highest agreement with the PAHO model across countries (>91.9%). Other NP with high agreement with the PAHO model were Ecuador (89.8%), Uruguay (82.5%), Chile Phase 3 (82.3%), and Peru Phase 2(84.2%). In contrast, Peru Phase 1, Brazil, and Chile Phase 1 NP models had the highest proportion of foods classified as healthy (49.2, 47.1 and 46.5%, respectively), and the lowest agreement with the PAHO model (69.9%, 69.3%, and 73% respectively).

Conclusions. The three implementation phases of the Mexican NP had a near to perfect agreement and a strong correlation with the PAHO model and were valid to identify healthy food products. In contrast, Brazil NP model had the lowest correlation with the PAHO model. The results of this study can help countries seeking to adapt and validate existing models for use in population-specific applications.