Pre-Meeting Workshop: Getting Started with Bayesian PK/PD Modeling and Simulation Using WinBUGS

To take part in the hands-on examples, all workshop attendees will need to install some tools on their laptop before arriving at the course. If you did not receive an email with instructions for downloading the material, please contact Heidi at heidic@metrumrg.com.

Date: Sunday March 9, 2008
Time: 8:00 am to 4:30pm

Instructor: William R. Gillespie, Ph.D. and Metrum Institute Staff

Objective: Provide an introduction to Bayesian modeling and the practical use of WinBUGS 1.4.2., WinBUGS scripting and the Pharmaco interface for PK/PD applications.

Intended Audience: Pharmacometrics scientists with experience in population PK/PD modeling and a basic understanding of nonlinear mixed effects modeling methodology.

Workshop Outline:

Introduction to Bayesian statistical principles and methods
- Bayes' Rule
- Bayesian modeling & inference process
  - Probability as a measure of one's state of knowledge
  - Contrast/compare with frequentist approach to probability and inference
- Likelihood principle
  - Practical implications
Computation for Bayesian modeling
- Key challenge of Bayesian modeling and inference: high-dimensional multiple integration
- General computational approach: posterior simulation
- Brief intro to Markov chain Monte Carlo simulation
WinBUGS basics
- What is it? How do I get it? How do I run it?
- Using the WinBUGS model language
- Entering data and initial estimates
- Generating MCMC samples
- Analyzing and interpreting the results
Topics in Bayesian model development
- Using WinBUGS scripts
- R tools for running WinBUGS and analyzing MCMC simulations
- Assessing convergence
- Programming hierarchical models (aka mixed effect or population models)
- PKBUGS and Pharmaco interface for population PK
Applications of Bayesian M&S in Drug Development
- An overview of the current and potential role of Bayesian methods in clinical drug development will be presented, including a brief look at applications to motivate interest in these approaches.
Hands-on examples:
Hands-on examples will be used to illustrate key concepts throughout the workshop. Participation in the hands-on exercises is optional and will require that attendees bring a laptop computer. All software used in the workshop is available by free download. Specific details about hardware/software requirements and installation instructions will be provided in advance of the workshop.
- Hands-on example 1: Simple nonlinear regression e.g., a PD model relating a single exposure metric to a single continuous PD outcome
- Hands-on example 2: Nonlinear mixed effects population PD model.
- Hands-on example 3: Nonlinear mixed effects population PK model.

Workshop Fee: $500 ($200 for students/postdoctoral fellows; number of student seats is limited). Fee includes electronic copies of course materials, code and scripts for hands-on examples, continental breakfast, breaks, and lunch. Total workshop is limited to 50 registrants.