Ultra-processed food consumption modifies the association of Nutri-Score with all-cause mortality. (20th October 2021)

• Record Type: Journal Article
• Authors: Bonaccio, M; Di Castelnuovo, A; Ruggiero, E; Costanzo, S; Esposito, S; Persichillo, M; Cerletti, C; Donati, MB; de Gaetano, G; Iacoviello, L
• Is Part Of: European journal of public health. Volume 31(2021)Supplement 3
• Publication Date: 2021-10-20
• Subjects: Epidemiology -- Europe -- Periodicals; Public health -- Europe -- Periodicals; 362.109405
• DOI: 10.1093/eurpub/ckab164.448
• Languages: English
• ISSNs: 1101-1262

Abstract

Abstract: Background: The Nutri-Score front-of-pack label rates foods according to their nutritional quality while the NOVA classification is focused on degree of processing; both systems separately predicted mortality in epidemiological contexts. We evaluated whether ultra-processed food (UPF) modifies the association of the Nutri-Score with mortality. Methods: Longitudinal analysis on 22, 549 Italian men and women (mean age 54±12 y) from the Moli-sani Study (2005-2010) followed for 11.2 y. Food intake was assessed by a food frequency questionnaire. The Food Standards Agency nutrient profiling system (FSAm-NPS) score, used to derive the Nutri-Score, was calculated for each food based on its amount of energy, saturated fat, sugar, sodium, fibre, protein, fruits, vegetables, legumes and nuts per 100 g of product. The FSAm-NPS dietary index was calculated for each participant as an energy weighted mean of the FSAm-NPS score of all foods consumed, and increased values indicate lower nutritional quality. UPF intake was the proportion (%) of UPF in the total weight of food consumed (g/d) and categorized as low/high (below/above the population median, respectively). Results: In a multivariable analysis, 1-SD increase in the FSAm-NPS index led to increased risk of all-cause (HR = 1.07; 95%CI 1.02-1.13) and CVD mortality (HR = 1.08; 1.00-1.18). UPF intake was an effect modifier since the association of FSAm-NPS index with mortality was confined to individuals with high UPF intake (HR = 1.14; 1.05-1.25) but not in those with low UPF (HR = 1.00; 0.93-1.07; p for interaction=0.034). Similar findings were reported for CVD mortality (HR = 1.14;0.99-1.31 and HR = 1.01;0.90-1.13 for high and low UPF consumers, respectively; p for interaction=0.18). Conclusions: Higher mortality risk associated with a NutriScore revealing nutrient-poor food intake is limited to individuals also reporting high UPF intake. Food labelling systems should account for food processing in addition to nutritional content. Key messages: Ultra-processed food modifies the association between the 5-color Nutri-Score front-of-pack label with all-cause mortality in an Italian general population. Food labelling systems, that are conceived to help consumers make healthier food choices, should also account for food processing in addition to nutritional content.