Gordon Graham
Centre for Applied Pharmacokinetic Research, School of Pharmacy and Pharmaceutical Sciences, University of Manchester, Oxford Road, Manchester, M13 9PL, UK.
Design issues of pharmacokinetic/pharmacodynamic studies have regained interest in recent years with simulation techniques of clinical trial designs being the main focus (Holford et al 1999). D-optimality has also received some attention as attempts have been made to generalise optimal design to random effects models (Metre et al 1997). In the statistics literature, Bayesian optimal design has also advanced with a (Markov chain) Monte Carlo approach (Muller and Parmigiani 1995). This latter method potentially leads to the implementation of optimal design for any type of model and experimental complexity.
This work considers a Bayesian decision theoretic framework that allows a general approach for the design of experiments using the method of Muller and Parmigiani (1995). The utility function of interest is the Kullback-Leibler distance proposed by Lindley (1956) as a measure of information provided by an experiment. This is asymptotically equivalent to the D-optimal design of an experiment resulting in a linear model with vague prior parameter information and the Bayesian D-optimal design for a nonlinear model.
Two simple models are considered here: the one compartment first order absorption model and the logistic regression model. Bayesian optimal design is applied to these models and the results are compared to known results using D-optimal and Bayesian D-optimal design.
References:
N.H.G. Holford, M. Hale, H.C. Ko, J.-L. Steimer, L.B. Sheiner, C.C. Peck (Editors). Simulation in drug development: Good practices. Publication of CDDS, http://www.dml.georgetown.edu/cdds/sddgp723.html (1999).
D.V. Lindley. On a measure of the information provided by an experiment. Ann. Math. Statist., 27: 986-1005 (1956).
F. Mentre, A. Mallet, D. Baccar. Optimal design in random-effects regression models. Biometrika, 84: 429-442 (1997).
P. Muller, G. Parmigiani. Optimal design via curve fitting of Monte Carlo experiments. J. Am. Stat. Ass., 90: 1322-1330 (1995).
Reference: PAGE 10 () Abstr 178 [www.page-meeting.org/?abstract=178]
Poster: poster