Background. Digoxin reduces hospitalizations due to heart failure (HF) and may also reduce mortality at low serum digoxin concentrations (SDC). Most HF patients are ≥ 65 years, yet the effects of digoxin on outcomes in these patients have not been well studied. Methods. Of the 7788 ambulatory chronic HF patients in normal sinus rhythm in the Digitalis Investigation Group trial (1991-1995), 5548 (2890 were ≥ 65 years) were alive at 1 month and were either receiving placebo or had data on SDC. Of these patients, 982 had low (0.5-0.9 ng/mL) and 705 had high (≥ 1 ng/mL) SDC. Results. Among patients ≥ 65 years, compared with 38% placebo patients, 34% low SDC patients died during 39 months of median follow-up (adjusted hazard ratio [AHR] = 0.81; 95% confidence interval [CI] = 0.68-0,96; p = .017). All-cause hospitalizations occurred in 70% of placebo and 68% of low-SDC patients (AHR = 0.86; 95% CI = 0.76-0.98; p = .019). Reduction in hospitalizations for HF occurred in both low and high SDC groups. High SDC was not independently associated with all-cause hospitalization or all-cause mortality. Age, impaired renal function, and pulmonary congestion reduced the odds of low SDC. Low-dose digoxin (≤ 0.125 mg/d) was the strongest independent predictor of low SDC (adjusted odd ratio = 2.37; 95% CI = 1.65-3.39); p < .0001). Conclusions. Digoxin at low SDC was associated with a reduction in mortality and hospitalization in chronic geriatric HF, and low-dose digoxin was the strongest predictor of low SDC. Copyright 2007 by The Gerontological Society of America.