Yale University
Department of Statistics

Mixture Models for Longitudinal Outcomes which Depend on Event Times


Monday, October 21, 1996

Joseph Hogan
Department of Statistics
Brown University
Seminar to be held in Room 309, LEPH, 60 College

Many long-term studies collect both a vector of repeated measurements and an event time on each subject; often, the two outcomes are dependent. One example is the use of surrogate markers to predict disease onset or survival. Another is a longitudinal clinical trial which has outcome-related dropout. We describe a mixture model for the joint distribution which accommodates incomplete repeated measures and right-censored event times, and provide methods for maximum likelihood estimation. Two applications are outlined: one is a randomized clinical trial of a new therapy for schizophrenia in which poor responders are systematically removed. The other is a pediatric AIDS trial in which an intention to treat analysis is recovered from incomplete data.