The goal of this study is to investigate the population pharmacokinetics of oral given clozapine in Chinese schizophrenic patients and to identify possible relationships between population parameters and covariates including demography factors and CYP1A2 genetic polymorphism, so as to create the population pharmacokinetics model to guide individual clinical delivery. Details of drug dosage history, sampling time and concentration of 626 data points from 183 patients were collected retrospectively. The 183 patients were randomly allocated either to the index group (n = 168) or to the validation group (n = 15). Population pharmacokinetic data analysis was performed using the nonlinear mixed-effects model (NONMEM) program on the index group. The values of apparent clearance (CL/F), apparent volume of distribution (V/F) and the constant of absorption rate were estimated. A number of covariates including demographic index, coadministration of other drugs and CYP1A2 genotypes were evaluated statistically for their influence on these parameters. The final population model related clearance with day-dose/BSA (DBSA) and smoke habit (SMOK). Predictive performance of the final model evaluated with the validation group showed insignificant bias between observed and model predicted concentrations. Typical value of CL/F (non-smoking group), V/F and the constant of absorption rate were 28.5 L x h(-1) (5.05%), 1 290 L (16.7%) and 2.26 h(-1) (fixed), inter-patient variability (CV) in CL/F and V/F was) 42.2% and 10.0%, respectively. It was observed that the values of CL/F in the two smoking groups were higher than that in the non-smoking group. The residual variability (SD) between observed and model-predicted concentrations was 45.8 microg x L(-1).