Purpose: This study aimed to utilize a population pharmacokinetic method to obtain information about the influence of covariates on the in vivo behavior of digoxin in patients with cardiac insufficiency. Methods: A total of 228 therapeutic drug monitoring concentrations were retrospectively collected from 176 inpatients. The patients were randomly divided into a modeling group (n = 126) and a validation group (n = 50). The first-order absorption one-compartment model was used to develop a population pharmacokinetic model from a nonlinear mixed effects modeling approach. Sixteen single nucleotide polymorphisms involved in the pharmacokinetic variables of digoxin were identified by using the MassARRAY system. Various demographic parameters, biochemical test values, concomitant medications, and genetic variants were investigated. Findings: The typical population value of digoxin CL/F was 5.06 L/h, and the volume of distribution was 211.82 L. The drug CL/F was significantly related to serum creatinine, in a combination of spironolactone and SLCO4C1 genotypes of 2 variants (rs3114660 and rs3114661). Results of model evaluation and internal/external validation indicated a stable and precise performance of the final model. Implications: For the first time, 2 single nucleotide polymorphisms (rs3114660 and rs3114661) in SLCO4C1 were found to significantly affect the elimination of digoxin in vivo. The final population model may be useful for the individualized dosing of digoxin for patients with cardiac insufficiency. (c) 2020 Elsevier Inc.