BACKGROUND: Excess sodium intake is associated with high blood pressure, a major risk factor for cardiovascular disease (CVD). It is unknown whether decreasing sodium intake to <2300 mg/d has an effect on CVD or all-cause mortality. OBJECTIVE: The objective was to assess whether reductions in sodium intake to <2300 mg/d were associated with either an increased or a decreased risk of fatal and nonfatal CVD and all-cause mortality. DESIGN: This observational prospective study of the PREvención con DIeta MEDiterránea (PREDIMED) trial included 3982 participants at high CVD risk. Sodium intake was evaluated with a validated food-frequency questionnaire and categorized as low (<1500 mg/d), intermediate (≥1500 to ≤2300 mg/d), high (>2300 to ≤3400 mg/d), or very high (>3400 mg/d). Subsequently, 1-y and 3-y changes in sodium intake were calculated. Multivariate relative risks were assessed by using Cox proportional hazards ratios. Marginal structural models with inverse probability weighting were used to test the effect of changes in sodium intake and the Mediterranean diet (MedDiet). RESULTS: We documented 125 CVD events and 131 deaths after a 4.8-y median follow-up. Sodium intake <2300 mg/d was associated with a lower risk of all-cause mortality: 48% (HR: 0.52; 95% CI: 0.30, 0.91; P = 0.02) and 49% (HR: 0.51; 95% CI: 0.26, 0.98; P = 0.04) after 1 and 3 y, respectively. Increasing sodium intake after 1 y was associated with a 72% (HR: 1.72; 95% CI: 1.01, 2.91; P = 0.04) higher risk of CVD events. The incidence rate of CVD was reduced for those who reduced their sodium intake and were randomly assigned to MedDiet interventions [4.1/10,000 (95% CI: 3.1, 8.0) compared with 4.4/10,000 (95% CI: 2.7, 12.4) person-years; P = 0.002]. CONCLUSIONS: Decreasing sodium intake to <2300 mg/d was associated with a reduced risk of all-cause mortality, whereas increasing the intake to >2300 mg/d was associated with a higher risk of CVD. Our observational data suggest that sodium intake <2300 mg/d was associated with an enhanced beneficial effect of the MedDiet on CVD. These results should be interpreted with caution, and other confirmatory studies are necessary.