PURPOSE:To prospectively assess the association of dietary fiber intake (from different dietary sources) with all-cause mortality in a Mediterranean cohort. METHODS:We assessed 19,703 participants of the SUN (Seguimiento Universidad de Navarra) cohort (mean follow-up: 10.1 years). A validated 136-item FFQ was administered at baseline. We used Cox proportional hazards models adjusted for multiple socio-demographic, anthropometric, lifestyle factors, and prevalent conditions at baseline. RESULTS:We observed 323 deaths during 198,341 person-years of follow-up. A significantly inverse linear trend in Cox models was observed for the association of total dietary fiber intake and all-cause mortality after adjustment for confounders (p for trend 0.017). Each additional intake of 5 g/1000 kcal of dietary fiber was associated with a 9% relative reduction in all-cause mortality risk (HR 0.91, 95% CI 0.84-0.99). Considering separate dietary sources in separate models, a significant inverse trend was apparent for fiber derived from vegetables (p for trend 0.001), but it was non-significant for fiber derived from fruit, legumes, cereals, or other sources. Soluble fiber was significantly inversely associated with all-cause mortality in the fully adjusted model (p for trend 0.007), and insoluble fiber was marginally significant (p for trend 0.08). CONCLUSIONS:A higher intake of total dietary fiber, and particularly fiber from vegetables, was related to a reduced all-cause mortality in our Mediterranean cohort. Dietary messages to increase the consumption of dietary patterns rich in fiber-rich foods should be broadly disseminated to decrease the alarming rate of chronic diseases and its derived mortality.