Abstract. Taranabant is a cannabinoid-1 receptor inverse agonist developed for the treatment of obesity. A population model was constructed to facilitate the estimation of pharmacokinetic parameters and to identify the influence of selected covariates. Data from 12 phase 1 studies and one phase 2 study were pooled from subjects administered single and multiple oral doses of taranabant ranging from 0.5 to 8 mg. A total of 6,834 taranabant plasma concentrations from 187 healthy and 385 obese subjects were used to develop the population model in NONMEM. A standard covariate analysis using forward selection (α=0.05) and backward elimination (α=0.001) was conducted. A three-compartment model with firstorder absorption and elimination adequately described plasma taranabant concentrations. The population mean estimates for apparent clearance and apparent steady-state volume of distribution were 25.4 L/h and 2,578 L, respectively. Statistically significant covariate effects were modest in magnitude and not considered clinically relevant (the effects of body mass index (BMI) and creatinine clearance (CrCL) on apparent clearance; BMI, age, CrCL, and gender on apparent volume of the peripheral compartment and age on apparent intercompartmental clearance). The pharmacokinetic profile of taranabant can adequately be described by a three-compartment model with first-order absorption and elimination. Clinical dose adjustment based on covariates effects is not warranted.