Association Between Dietary Spermidine Intake and Mortality in the Population-based Bruneck Study

This prospective community-based cohort study includes 829 participants aged 45-84 years, 49.9% of which are male. Diet is assessed by repeated dietician-administered validated food-frequency questionnaires (2540 assessments) in 1995, 2000, 2005, and 2010. Nutrient intakes including spermidine are calculated using USDA standard databases. Clinical events (all-cause mortality and cause-specific mortality) are recorded from 1995 to 2015.

The primary endpoint of all-cause mortality is related to the exposure of long-term average spermidine intake by Cox proportional hazards models with time-varying covariates. Additional analyses employ the Fine and Gray proportional subdistribution hazards model and flexible Royston-Parmar spline-based models. Comprehensive sensitivity analyses are performed to guard against potential biases associated with nutritional epidemiology.