From: "Nandy, Partha"
Subject: [NMusers] Population PK model for Incomplete Block Design
Date: 12/3/2003 11:40 AM

Hi All,

I would appreciate some input on how to establish a population
PK model for the following:

The study was a bioavailability study with an incomplete block
design, where each subject crossed over to receive 4 treatments
in all in a completely random order. Altogether four (4) formulations
were evaluated.  One was an IR (immediate release) formulation and
the other three were controlled release formulations.  The formulations
were also tested both under fasted or fed condition.  So, in all there
were total of 8 treatments.

Can some one suggest how to tackle this problem?  I started off with
estimating a whole bunch of THETAS, but ran into problem (as expected). 
I can break up the data set into smaller bits, but I was wondering if there
is a way around it.  I would really want to model all the treatments simultaneously.



(Partha Nandy)