Population Pharmacokinetic Parameter Estimates using a Limited Sampling Design: Analysis of Blood Alcohol Levels

Abstract

When population pharmacokinetic (PPK) analysis is conducted, there are often constraints on obtaining blood samples at multiple time points. In this context, it is a concern that estimated PPK parameters are reported without an evaluation of their reliability. The aim of the present study was to improve the reliability of the estimated population mean pharmacokinetic parameters after assessing the precision and accuracy of the estimates by stochastic simulation and estimation. In an example of alcohol metabolism data, where samples taken later than 60 min after consumption were not available, the precision of the estimates for the population mean elimination rate “constant” (kel) was found to be low. To tackle this problem, real-life data were aggregated with the data stochastically generated from historical knowledge, and the parameters were estimated. The estimated kel value based on only real data was revealed to be biased toward a lower value.