This study was aimed at determining the population pharmacokinetics of digoxin and identifying factors that explain pharmacokinetic variability in elderly patients. The data of 142 elderly patients and 448 samples were collected after repetitive oral digoxin. Blood samples were drawn at various times after administration. Population pharmacokinetic analysis was performed using nonlinear mixed effects modelling program (NONMEM). A one-compartment model with first-order absorption and elimination was selected as the base model. The influence of demographic characteristics, biochemical and haematological indices as well as other commonly used co-medications were explored. The typical values with interindividual variability for apparent clearance (CL/F) and apparent volume of distribution (V/F) were 8.9 L h(-1) (43.2 %) and 420 L (65.8 %), respectively. The residual variability was 31.6 %. CL/F decreased significantly with renal function, total body weight, calcium channel blockers or spironolactone co-therapy and symptom with congestive heart failure. The median parameter estimates from a nonparametric bootstrap procedure were comparable and within 5 % of the estimates from NONMEM. These results provide important information for clinicians to optimize digoxin regimens in elderly patients.