Estimation of Pooled Pharmacokinetic Parameters Describing Populations

Abstract

By population pharmacokinetics we mean the typical relationships between physiology and pharmacokinetics; the interindividual variability in these relationships, and their residual inexplicable intraindividual variability. Knowledge of such population features is useful both to gain insight into the drug-patient system, and to adjust individual drug dosage. Experimental data from which population kinetics might be estimated, often comes from only a few atypical individuals (e.g., normal volunteers, rather than patients) More representative data might be those coming from routine patients These data, however, are marked by varying quality, accuracy and precision, as well as there being little data per patient. To combine data from various sources, and to use routine data, one must overcome certain data analysis problems. The standard (experimental data oriented) approach does not do so. An approach is available that regards the population, rather than the individual, as the primary unit of analysis. It is useful for analysis of routine data and for combining data of varying quality. This paper discusses the approach and contrasts it with the standard one.