Ultra-processed food consumption, plasma metabolite profile, and risk of all-cause and cause-specific mortality in a population-based cohort

Background & aims: Epidemiological evidence on ultra-processed food (UPF) and cause-specific mortality remains limited and mixed. Molecular mechanisms underlying UPF intake and mortality remain unexplored. This study aimed to evaluate the associations between UPF consumption, metabolic signatures, and all-cause, premature, and cause-specific mortality. Methods: This study included 27670 participants (mean age 58.1 years) from the Malmö Diet and Cancer (MDC) cohort study. Consumption of UPF was assessed using a food frequency questionnaire and a 7-day food diary. In a subset of the MDC (n = 879), the associations of UPF with 991 plasma metabolites were investigated. An elastic net regression model was used to establish the metabolic signature of UPF. Cox proportional hazards regression model was used to determine the association between UPF intake, metabolic signature, and mortality risk. Results: During a median follow-up of 23.3 years, a total of 11333 participants died. UPF intake showed a nonlinear positive association with all-cause mortality, with more pronounced associations found in females (P_interaction = 0.044); in females, UPF was linked to a higher mortality risk in a linear manner, while the association was J-shaped in males. Each standard deviation (SD) increment in UPF intake was associated with an increased risk of premature mortality (HR, 1.06; 95 % CI, 1.03–1.09), cardiovascular disease (CVD) mortality (HR, 1.05; 95 % CI, 1.01–1.08) or respiratory disease mortality (HR, 1.08; 95 % CI, 1.01–1.15), but not cancer mortality. The metabolic signature for UPF consumption (with 93 metabolites) was positively associated with all-cause mortality risk (HR per 1 SD, 1.23; 95 % CI, 1.06–1.42). Conclusions: Our results suggest that higher UPF intake is associated with increased risk of all-cause, premature, CVD, and respiratory disease mortality, with the association varying across sex for all-cause mortality. The plasma metabolic signature of UPF showed a positive association with all-cause mortality.