Date: Tue, 08 Feb 2000 08:58:00 +0100
From: Huitema Alwin <APAHU@slz.nl>
Subject: Optimal sampling points

We have the following problem. We want to estimate individual PK parameters using a Bayesian approach with sparse individual data based on a prior population PK analysis. We have selected optimal sampling times using the D-optimality criterion as incorporated in ADAPT II using the typical values calculated from the population PK analysis. This criterion is, however, not optimal since it is designed when data are analyzed without reference to a prior distribution. Does anyone know how to select optimal sampling points for a Bayesian estimation procedure and how this is practically performed?

Alwin Huitema

 

 

*****

 

 

Date: Tue, 08 Feb 2000 08:21:27 -0800
From: LSheiner <lewis@c255.ucsf.edu>
Subject: Optimal sampling points

You might try loking at the work of Mentre & coworkers, J Pharmacokin Biopharmaceut 23: 101, 1995; Stat in Med 13: 185, 1994.

LBS.


--
_/ _/ _/_/ _/_/_/ _/_/_/ Lewis B Sheiner, MD (lewis@c255.ucsf.edu)
_/ _/ _/ _/_ _/_/ Professor: Lab. Med., Bioph. Sci., Med.
_/ _/ _/ _/ _/ Box 0626, UCSF, SF, CA, 94143-0626
_/_/ _/_/ _/_/_/ _/ 415-476-1965 (v), 415-476-2796 (fax)

Date: Wed, 09 Feb 2000 21:48:54 -0500


*****



From: Bill Gillespie <bgillespie@pharsight.com>
Subject: Re: Optimal sampling points

There is a body of literature on Bayesian optimal design and optimal sampling times would be a special case. Some recent work is described in:

P Müller. Simulation based optimal design. in JM Bernardo, JO Berger, AP Dawid, AFM Smith (eds.). Bayesian Statistics 6. Oxford Univ. Press (1998).

GL Rosner. Simulation-based methods of determining optimal sampling times. Presented at BMSR Modeling Workshop 1999.

Both authors are located at Duke ( http://www.stat.duke.edu/ ).

Bill Gillespie

Huitema Alwin wrote:

 

 

*****

 

 

Date: Tue, 15 Feb 2000 12:46:13 +0100
From: Huitema Alwin <APAHU@slz.nl>
Subject: Optimal sampling times

We have recently asked for methods for obtaining optimal sampling times for Bayesian estimation with knowledge of the prior distribution. We are aware of literature on this issue (e.g. Mentre in JPB), but we do not have any experience in applying these techniques. Is there any readily available software package in which these calculations are incorporated?

Alwin Huitema

 

 

*****

 

 

From: "Yuen, Geoffrey J" <gjy25826@glaxowellcome.com>
Subject: RE: Optimal sampling times
Date: Tue, 15 Feb 2000 10:04:57 -0500

The previous e-mails had listed a website at Duke University that details some of the software available (it's called BUGS).

Geoffrey Yuen, Pharm.D.
Clinical Pharmacokineticist
Glaxo-Wellcome
5 Moore Drive
RTP, NC 27709
phone 919 483 5676
email gjy25826@glaxowellcome.com