A Validation Study of NOVA Classification for Ultra-Processed Food on the USDA Food and Nutrient Database

Autor: Marco Silano, Rita Di Benedetto, Giulia Lorenzoni, Dario Gregori, Honoria Ocagli
Rok vydání: 2021
Předmět:
Zdroj: Curr Dev Nutr
ISSN: 2475-2991
DOI: 10.1093/cdn/nzab044_025
Popis: OBJECTIVES: In the last years, a group of Brazilian researchers has proposed a new food classification system, named NOVA, based on the extent of food processing. However, the feasibility of such classification has been debated, together with its cross-cultural validity. The present work assessed the NOVA classification feasibility, conducting a validation study on the USDA Food Composition Database. METHODS: Two independent reviewers rated each food reported in the 2015–2016 USDA Food and Nutrient Database to be or not ultra-processed food (UPF) according to the criteria presented in the manuscripts published by Monteiro CA et al. in 2016 and 2019 presenting the NOVA classification. A third independent reviewer solved disagreements. The Cohen's Kappa was calculated to evaluate the agreement between the two independent reviewers. RESULTS: The agreement between the reviewers was only moderate, with a Cohen's Kappa of 0.58. The disagreement pertained mainly the mixed dishes since it was difficult for the two independent reviewers to rate (UPF or not) the single food components of the mixed dishes. CONCLUSIONS: Such work provides insights on the difficulties encountered in applying the NOVA classification to a real-word food database outside the cultural context in which the NOVA classification was developed. FUNDING SOURCES: N/A.
Databáze: OpenAIRE