https://doi.org/10.37527/2023.73.S1
1Program of Collective Health, School of Medical Sciences, Universidade Estadual de Campinas, Campinas, Brazil, 2Center for Food Studies and Research, Universidade Estadual de Campinas, Campinas, Brazil, 3NUPENS, Universidade de São Paulo, São Paulo, Brazil.
Background and objective: Historical nationwide publicly available data on packaged food and beverage composition is not available in most countries, including Brazil, which presents a challenge in monitoring and evaluating food and nutrition policies. The objective of the study was to verify whether a commercial dataset of packaged foods and beverages sold in Brazil is reliable using as ‘gold standard’ primary data collected in the country. Methods: We used commercial data from the global database of packaged foods and beverages, Mintel GNPD. To analyze the reliability of these, we compared the average amounts of total fat, saturated fat, trans-fat, total sugar, and sodium per 100g or ml of products - considered nutrients of concern according to the PAHO nutritional profile model - with primary data of the 2017 Brazilian Food Database, collected in the five top retailers in Brazil. A total of 2496 foods and beverages from 2017 were found in both databases. We assessed the agreement using the paired t-test (p<0.05) and the Intraclass Correlation Coefficient (ICC>0.90) for each nutrient. Results: We found excellent agreement between the amount of total fat (ICC=0.99), saturated fat (ICC=0.94), trans fat (ICC=0.93), total sugar (ICC=0.99) and sodium (ICC=0.99) between primary data and commercial data. Furthermore, the averages of these nutrients did not significantly differ (p>0.05) between both databases to products analyzed. Conclusion: The commercial data on food and beverage content of targeted nutrients of concern was found to be reliable when compared with primary data collected in Brazil and are reliable to be used in monitoring and evaluating food and nutrition policies in the country.
Keywords: nutrition labeling; monitoring; evaluation; food policy; reliability.