Professor Richard J. Cook (Department of Statistics & Actuarial Science University of Waterloo)
will be giving the following seminar at the MRC Biostatistics Unit, Institute of Public Health,
on Tuesday 26th of September. The seminar starts at 4pm in the large seminar room.
Everyone is welcome.
Title: Efficiency and Design Considerations in Clinical Trials with Selection based
on Recurrent Events
Abstract
We describe a method for planning the duration of a randomized parallel
group study in which the response of interest is a potentially recurrent
event. We assume that patients are accrued at a constant rate, we model
events via a mixed time-homogeneous Poisson process, and we utilize an
independent exponential censoring mechanism to reflect loss to follow-up.
We derive the appropriate study duration to ensure power requirements are
met for the effect size of interest under a mixed Poisson regression model.
The impact of selecting patients for study based on responses measured
during a screening period is discussed in terms of bias and relative
efficiency of standard methods of analysis. Remarks are also made on the
implications dependent censoring due to a terminal event related to the
recurrent events.
Daniela De Angelis
MRC Biostatistics Unit Tel: + (0)1223 330390
Institute of Public Health
Robinson Way Fax: + (0)1223 330388
Cambridge
CB2 2SR Web: www.mrc-bsu.cam.ac.uk
United Kingdom
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|