ISPOR 11th Annual International Meeting May 20-24, 2006
Marriott Philadelphia, Philadelphia, PA
WORKSHOP PROPOSAL EXAMPLE
TIME-VARYING EFFECTS IN SURVIVAL ANALYSIS: TREATMENT EFFECTS IN DISGUISE
DISCUSSION LEADERS: Nicole Nitz MS, i3 Magnifi, Eden Prairie, MN, USA;
Sharashchandra Shetty PhD, i3 Magnifi, Eden Prairie, MN, USA;
Carolyn Harley PhD, i3 Magnifi, Eden Prairie, MN, USA
WORKSHOP PURPOSE: The purpose of this workshop is to help participants
better understand the issues related to the application of analysis
in outcomes research, with an emphasis on time-varying techniques in
isolating treatment effects.
WORKSHOP DESCRIPTION: The Cox Proportional Hazards model is one of the
most commonly used methods in time to event analysis. usefulness
of this technique depends upon whether researchers fully examine the
proportionality assumption. The average treatment effect
under the proportionality assumption may not accurately capture
differences between treatment cohorts that change over time. Failure
the possibility of non-proportionality therefore may lead to erroneous
conclusions. Using specific examples of health outcomes research workshop
will identify potential factors that may cause differences in treatment
effects and will review techniques to address non-proportionality.
Participants will be engaged in a critical comparison of each technique
and its impact on study results through the use of an applied example.
Discussion and questions will be solicited throughout the presentation
(rather than limited to a question and answer period at the end.of the
session).
The health outcomes that will be assessed in the workshop will include
treatment adherence, hospitalization and health events.
