Premature deaths attributable to the consumption of ultra-processed foods: a comparative assessment modelling study in eight countries

ABSTRACT Background Ultra-processed foods (UPFs) are becoming dominant in the global food and supply. Prospective cohort studies have found an association between UPF dietary pattern and increased risk of several non-communicable diseases and all-cause mortality. In this study, we (1) estimated the risk of all-cause mortality associated for each 10% increase in the share of UPF consumption in the total energy intake; (2) estimated the population attributable fractions (PAF) and the total number of premature deaths attributable to the consumption of UPF in adults (30-69 years) from 8 selected countries. Methods First, we performed a dose-response meta-analysis of observational cohort studies assessing the association between UPFs dietary pattern and all-cause mortality. As we found evidence of linearity, we estimated the pooled RR (and its 95% CI) for all-cause mortality per each 10% increment in the % UPF. Then, we estimated the population attributable fraction (PAF) of premature all-cause mortality attributable to UPF in 8 selected countries with relatively low (Colombia and Brazil), intermediate (Chile and Mexico), and high (Australia, Canada, UK, and US) UPF consumption. Results We found a linear dose-response association between UPF intake and all-cause mortality, with a 2.7% increased risk of all-cause mortality per 10% increase in the % UPF. Considering the magnitude of the association between UPFs intake and all-cause mortality, and the dietary share of UPF in each of the 8 selected countries, we estimated that 4% (Colombia) to 14% (United Kingdom and United States) of premature deaths were attributable to UPF intake. Conclusions Our findings support that UPF intake contributes significantly to the overall burden of disease in many countries and its reduction should be included in national dietary guideline recommendations and addressed in public policies.