Abstract
In this paper 3 criteria to design experiments for Bayesian estimation of the parameters of nonlinear models with respect to their parameters, when a prior distribution is available, are presented: the determinant of the Bayesian information matrix, the determinant of the preposterior covariance matrix, and the expected information provided by an experiment. A procedure to simplify the computation of these criteria is proposed in the case of continuous prior distributions and is compared with the criterion obtained from a linearization of the model about the mean of the prior distribution for the parameters. This procedure is applied to two models commonly encountered in the area of pharmacokinetics and pharmacodynamics: the one-compartment open model with bolus intravenous single-dose injection and theE max model. They both involve two parameters. Additive as well as multiplicative gaussian measurement errors are considered with normal prior distributions. Various combinations of the variances of the prior distribution and of the measurement error are studied. Our attention is restricted to designs with limited numbers of measurements (1 or 2 measurements). This situation often occurs in practice when Bayesian estimation is performed. The optimal Bayesian designs that result vary with the variances of the parameter distribution and with the measurement error. The two-point optimal designs sometimes differ from the D-optimal designs for the mean of the prior distribution and may consist of replicating measurements. For the studied cases, the determinant of the Bayesian information matrix and its linearized form lead to the same optimal designs. In some cases, the pre-posterior covariance matrix can be far from its lower bound, namely, the inverse of the Bayesian information matrix, especially for theE max model and a multiplicative measurement error. The expected information provided by the experiment and the determinant of the pre-posterior covariance matrix generally lead to the same designs except for theE max model and the multiplicative measurement error. Results show that these criteria can be easily computed and that they could be incorporated in modules for designing experiments.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
J. L. Steimer, A. Mallet, and F. Mentré. Estimating interindividual pharmacokinetic variability. In M. Rowland, L. B. Sheiner, and J. L. Steimer (eds.),Variability in Drug Therapy: Description, Estimation and Control, Raven Press, New York, 1985, pp. 65–111.
J. L. Steimer, M. E. Ebelin, and J. Van Bree. Pharmacokinetic and pharmacodynamic data and models in clinical trials.Eur. J. Drug. Metab. Pharmacokin. 18:61–76 (1993).
A. H. Thomson and B. W. Whiting. Bayesian parameter estimation and population pharmacokinetics.Clin. Pharmacokin. 22:447–467 (1992).
E. Walter and L. Pronzato. Qualitative and quantitative experiment design for phenomenological models—A survey.Automatica 26:195–213 (1990).
V. V. Fedorov.Theory of Optimal Experiments, Academic Press, New York, 1972.
D. M. Steinberg and W. G. Hunter. Experimental design: Review and comment.Technometrics 26:71–97 (1984).
A. C. Atkinson. Developments in the design of experiments.Int. Statist. Rev. 50:161–177 (1982).
R. C. St John and N. R. Draper. D-Optimality for regression designs: A review.Technometrics 17:15–23 (1975).
H. J. Rostami. Experimental design for pharmacokinetic modeling.Drug Inform. J. 24:299–313 (1990).
I. Verdinelli. Advances in experimental design. InBayesian Statistics (4), Oxford University Press, Oxford, 1992, pp. 467–481.
H. N. Bandemer, N. Wolfgang, and J. Pilz. Once more: optimal experimental design for regression models.Statistics 18:171–217 (1987).
I. Verdinelli and J. B. Kadane. Bayesian designs for maximizing information and outcome.J. Am. Statist. Assoc. 87:510–515 (1992).
K. Chaloner. Optimal Bayesian experimental design for linear models.Ann. Statist..12:283–300 (1984).
C. E. Shannon. A mathematical theory of communication.Bell System Tech. J. 27:379–423, 623–656 (1948).
D. V. Lindley. On the measure of information provided by an experiment.Ann. Math. Statist. 13:986–1005 (1956).
Y. Merlé, F. Mentré, A. Mallet, and A. Aurengo. Designing an optimal experiment for Bayesian estimation: Application to the kinetics of iodine thyroid uptake.Statist. Med. 13:185–196 (1994).
Y. Merlé, F. Mentré, A. Mallet, and A. Aurengo. Computer-assisted individual estimation of radioiodine uptake in Grave's disease.Comput. Meth. Prog. Biomed. 40:33–41 (1993).
N. R. Draper and G. W. Hunter. The use of prior distributions in the design of experiments for parameter estimation in non-linear situation: Multiresponse case.Biometrika 54:662–665 (1968).
J. Pilz.Bayesian Estimation and Experimental Design in Linear Regression Models, Wiley, New York, 1991.
D. Katz. Discrete approximations to continuous density functions that are L1 optimal.Comput. Statist. Data Anal. 1:175–181 (1983).
D. Katz. Discrete approximation of multivariate densities with application to Bayesian estimation.Comput. Statist. Data Anal. 2:27–36 (1984).
A. Iliadis, A. C. Brown, and M. L. Huggins. APIS: A software for model identification, simulation and dosage regimen calculations in clinical and experimental pharmacokinetics.Comput. Meth. Prog. Biomed. 38:227–239 (1992).
L. Pronzato and L. E. Walter. Robust experiment design via stochastic approximation.Math. Biosci. 75:103–120 (1985).
F. Mentré, A. Mallet, J. L. Steimer, and F. Lokiec. An application of the population pharmacokinetics to the clinical use of cyclosporine in bone marrow transplant patients.Transpl. Proc. 20:466–470 (1988).
Y. Hashimoto and L. B. Sheiner. Designs for population pharmacodynamics: value of pharmacokinetic data and population analysis.J. Pharmacokin. Biopharm. 19:333–353 (1991).
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Merlé, Y., Mentré, F. Bayesian design criteria: Computation, comparison, and application to a pharmacokinetic and a pharmacodynamic model. Journal of Pharmacokinetics and Biopharmaceutics 23, 101–125 (1995). https://doi.org/10.1007/BF02353788
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/BF02353788