Performance of an iterative two-stage bayesian technique for population pharmacokinetic analysis of rich data sets

Johannes H. Proost*, Douglas J. Eleveld

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

48 Citations (Scopus)


Purpose. To test the suitability of an Iterative Two-Stage Bayesian (ITSB) technique for population pharmacokinetic analysis of rich data sets, and to compare ITSB with Standard Two-Stage (STS) analysis and nonlinear Mixed Effect Modeling (MEM).

Materials and Methods. Data from a clinical study with rapacuronium and data generated by Monte Carlo simulation were analyzed by an ITSB technique described in literature, with some modifications, by STS, and by MEM (using NONMEM). The results were evaluated by comparing the mean error (accuracy) and root mean squared error (precision) of the estimated parameter values, their interindividual standard deviation, correlation coefficients, and residual standard deviation. In addition, the influence of initial estimates, number of subjects, number of measurements, and level of residual error on the performance of ITSB were investigated.

Results. ITSB yielded best results, and provided precise and virtually unbiased estimates of the population parameter means, interindividual variability, and residual standard deviation. The accuracy and precision of STS was poor, whereas ITSB performed better than MEM.

Conclusions. ITSB is a suitable technique for population pharmacokinetic analysis of rich data sets, and in the presented data set it is superior to STS and MEM.

Original languageEnglish
Pages (from-to)2748-2759
Number of pages12
JournalPharmaceutical Research
Issue number12
Publication statusPublished - Dec-2006


  • Bayesian analysis
  • data analysis
  • mixed effect modeling
  • Monte Carlo simulation
  • population pharmacokinetics

Cite this