ACoP Post-Meeting Workshop: Binary and Categorical Data Modeling, Analysis and Simulation using NONMEM®

Sponsored by ICON Development Solutions

To register for this workshop, visit the information and registration page on the ICON website.

Date and Times: Wednesday March 12 (12.30pm - 6:30pm) - Thursday March 13 (9am - 5pm)

Instructor: Adrian Dunne, University College, Dublin

Overview: This lecture style course will introduce the concepts, including the basic statistical concepts involved in modeling, analyzing and simulating binary and categorical data. Both individual subject and population situations will be covered.

The emphasis will be applied, rather than theoretical, with the use of many data example sets and case studies. The methods described will be implemented and demonstrated in great detail using NONMEM and PDx-POP software programs.

During the sessions, the computations will be run live. Each participant will be provided with a copy of the data and NONMEM control files, in order to practice the examples after the course.

Workshop Fee: $2000 ($1600 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 20 registrants.

Workshop Agenda:

Day 1

12 :30 - 1 :30 Lunch

1 :30 - 3 :00 Lecture 1

Introduction to set out aims and general approach.
Review of Statistical inference.
Types of data - quantitative & categorical.
Binary data - Bernoulli and Binomial distributions.
Estimation - Maximum Likelihood, Least Squares and REML.

3:00 - 3:15 Break

3.15 - 4:45pm Lecture 2

Quality of the estimates.
Review of Sampling distributions.
Precision, Standard Error and Confidence Intervals.
Dependence on sample size.
Modeling continuous and binary data.
Use of transformations - Logit, Probit, Log-log & Complementary log-log.
Interchanging the categories.

4:45 - 5:00 Break

5:00 - 6.30 Lecture 3

Logistic Regression Model.
Empirical logits.
Bioassay - ED50 estimation.
Pharmacokinetic-Pharmacodynamic modeling with a binary response.
Goodness of fit - Predicted Values.

Day 2

8:15 - 9:00 Breakfast

9:00 - 10:30 Lecture 4

Analysis of Deviance
Comparing logistic models - Likelihood Ratio test.
Parallel line bioassay - Potency estimation.
Pharmacokinetic-Pharmacodynamic models.
Model checking - residuals, goodness-of-link test.

10:30 - 11:00 Break

11:00 - 12:30 Lecture 5

Model checking - overdispersion.
Mixed effects models.
Latent variables and Tolerance distribution.

12:30 - 1:30 Lunch

1 :30 - 3 :00 Lecture 6

Ordinal data.
Cumulative Logits, Continuation-Ratio Logits, Adjacent-Categories Logits..
Proportional Odds and Hazards Models.
Reversibility and Invariance.
Testing for Proportional Odds.
Model Comparison.
Mixed effects Pharmacokinetic-Pharmacodynamic modeling with an ordinal response.

3:00 - 3:30 Break

3:30 - 5:00 Lecture 7

Simulating ungrouped binary data.
Simulating grouped binary data.
Simulating binary data with random effects.
Simulating ordinal data.
Simulating Pharmacokinetic-Pharmacodynamic data with binary and ordinal responses and random effects.
General Discussion