Kinetic Bayesian estimation

Unfortunately, there has been minimal work on identifiability issues with respect to population kinetic analysis. The current work on identifiability of kinetic models (Jacquez, 1985 belli and DiStefano, 1980 Cobelli and Saccomani, 1990) focuses on estimation for an individual experiment. With regard to population analysis and Bayesian estimation of parameters for an individual from the population, it is the population analysis step for which identifiability issues need to be considered. Once a prior distribution... 

Taright, N., Mentre, E, Mallet, A., and Jouvent, R. (1994) Nonpar-ametric estimation of population characteristics of the kinetics of lithium from observational and experimental data individualization of chronic dosing regimen using a new bayesian approach. Ther Drug Monit 16 258-269. 

This ability is available in many software programs. NONMEM (Iconus, EUicott City, MD) has been widely used to estimate population models arising from both sparse and intensely sampled data. Other programs include WinNonMix (Pharsight Corp., Palo Alto, CA), Kinetica 2000 (Innaphase Corp, Philadelphia, PA), and Pop-Kinetics (SAAM Institute, Seattle, WA). ADAPT II and WinNonlin have focused on PK/PD models and have been combined with Bayesian approaches to estimate population models. 

---