Paul,
You can include all measurements using one ID variable. However, you may want to include other indicator variables in the dataset to test for differences in PK with respect to period of pregnancy (i.e., EARLY=1 if sample during early preg. and =0 otherwise, LATE=1 if during late preg. and =0 otherwise, POST=1 if during post preg. and =0 otherwise).
To get an idea about whether pregnancy stage has an impact on the PK, you can look at a plot of WRES vs. Time Since First Dose with different symbols/colors for early,late and post. If you want to examine the potential effect a little more formally, you could also define inter-occasion variability (using etas) for the PK parameters using early, late, and post as the occasions (This will assign multiple individual values of the PK parameters for each subject). This approach will allow you to look at box plots of the PK parameters vs. pregnancy stage. The graphs can help you obtain reasonable initial estimates for testing pregnancy stage as a predictor of the appropriate PK variables. I would recommend testing the significance of the covariate with and without inter-occasion variability. If the preg. stage is an informative predictor, it could cause the IOV based upon preg. stage to become negligible and lead to difficulties with model minimization.
The eta/omega will then capture variability between individuals and sigma the remaining residual variability (intra-individual, assay error, time error, etc.)
Hope all goes well,
Luann Phillips
Director, PK/PD
Cognigen Corporation
[email protected] wrote:
> Hi,
>
> Recently we are modeling the ranitidine data in pregnant women. Some of
> them were sampled several times in different periods, for example, early,
> late or post pregnancy. Should we put all the samplings under one ID of
> the patient? or we can treat the several samplings as independent
> samplings and give them different ID? Anyone has such repeated measures
> experiences and any suggestions will be great appreciated.
>
> Paul
>
> School of Pharmacy
> University of Pittsburgh
> 716 Salk Hall
> 3501 Terrace Street
> Pittsburgh, PA 15261
> Phone: 412-648-8546
> E-mail: [email protected]