Background: The effect of loop diuretics on clinical outcomes in heart failure has not been evaluated in randomized controlled trials. In hospitalized patients with heart failure, a discharge loop diuretic prescription has been shown to be associated with improved 30-day outcomes, which appears to be more pronounced in subgroups with congestion. In the current study, we examined these associations and association modifications during longer follow-up. Methods: We assembled a propensity score-matched cohort of 2191 pairs of hospitalized heart failure patients discharged with, vs without, a prescription for loop diuretics, balanced on 74 baseline characteristics (mean age 78 years; 54% women; 11% African American). Results: Hazard ratio (HR) and 95% confidence interval (CI) for 6-year combined endpoint of heart failure readmission or all-cause mortality was 1.02 (0.96-1.09). HRs and 95% CIs for this combined endpoint in patients with no, mild-to-moderate, and severe pulmonary rales were 1.19 (1.07-1.33), 0.95 (0.86-1.04), and 0.77 (0.63-0.94), respectively (P for interaction, <.001). Respective HRs (95% CIs) for no, mild-to-moderate, and severe lower extremity edema were 1.16 (1.06-1.28), 0.94 (0.85-1.04), and 0.71 (0.56-0.89; interaction P <.001). Conclusions: The association between a discharge loop diuretic prescription and long-term clinical outcomes in hospitalized patients with heart failure is modified by admission congestion with worse, neutral, and better outcomes in patients with no, mild-to-moderate, and severe congestion, respectively. If these findings can be replicated, congestion may be used to risk-stratify patients with heart failure for potential optimization of loop diuretic prescription and outcomes.