Discontinuation of digoxin is associated with worsening heart failure (HF) symptoms. However, the long-term effects of discontinuation of digoxin therapy on mortality and morbidity in HF have not been well studied. Of the 7,788 participants in the Digoxin Investigation Group trial, 3,365 received digoxin before randomization. During the trial, digoxin was continued in 1,666 patients and discontinued in 1,699 patients. Using multivariable Cox regression analyses, we first determined the effect of discontinuation of digoxin on mortality and hospitalization during 39.7 months of median follow-up. Of the 1,666 patients continued on digoxin, 457 had low (0.5 to 0.9 ng/ml) and 340 had high (≥1.0 ng/ml) serum digoxin concentrations (SDC) after 1 month of therapy and of the 1,699 patients whose digoxin was discontinued, 1,674 were alive at 1 month. We examined the effects of continuation of digoxin at low or high SDC. Compared with continuation of long-term digoxin therapy, discontinuation of digoxin was associated with a significant increase in all-cause hospitalization (adjusted hazard ratio [AHR] 1.18, 95% confidence interval [CI] 1.09 to 1.28, p <0.0001) and HF hospitalization (AHR 1.35, 95% CI 1.20 to 1.51, p <0.0001), but had no effect on all-cause mortality (AHR 1.06, 95% CI 0.95 to 1.19, p = 0.272). In contrast, continuation of digoxin at low SDC was associated with a reduction in all-cause mortality (AHR 0.75, 95% CI 0.63 to 0.90, p = 0.002), all-cause hospitalization (AHR 0.80, 95% CI 0.70 to 0.91, p = 0.001), and hospitalization for HF (AHR 0.60, 95% CI 0.50 to 0.73, p <0.0001). In conclusion, continuation of long-term digoxin therapy at low SDC was associated with reduction in mortality and hospitalization in ambulatory patients with chronic HF receiving background therapy with angiotensin-converting enzyme inhibitors and diuretics. © 2007 Elsevier Inc. All rights reserved.