The population approach to pharmacokinetic data analysis: rationale and standard data analysis methods

Population pharmacokinetics describe the typical relationships between physiology (both normal and disease altered) and pharmacokinetics, the interindividual variability in these relationships, and their residual intraindividual variability. Knowledge of population kinetics can help one to choose initial drug dosage, to modify dosage appropriately in response to observed drug levels, to make rational decisions regarding certain aspects of drug regulation, and to investigate and elucidate certain research questions in pharmacokinetics. Experimental data from which population kinetics might be estimated often come from only those few individuals both willing and available to be studied. Clinical data from patients undergoing care might be more representative. These data, however, are marked by varying quality, accuracy, and precision, as well as there being few data per patient. Population pharmacokinetic parameters have traditionally been estimated either by fitting all individuals' data together as though there were no individual kinetic differences [the naive pooled data (NPD) approach], or by fitting each individual's data separately and then combining the individual parameter estimates [the two-stage (TS) approach]. These methods have certain theoretical problems which can only be aggravated when the deficiencies of data typical of clinical data are present. In this paper, the standard approaches are discussed and illustrated (using nondeficient data) in order to introduce subsequent papers in which alternative data analysis methods for population parameter estimation are defined, discussed, and compared.