|
Activity Number:
|
38
- Study Design: Historical Controls, Endpoint Timing, and Regulatory Considerations
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, July 29, 2018 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Medical Devices and Diagnostics
|
|
Abstract #330463
|
Presentation
|
|
Title:
|
An Adaptive Test of Significance in the Presence of Uncertainty in the Timing of the Final Analysis
|
|
Author(s):
|
Jeremy Gorelick*
|
|
Companies:
|
Edwards Lifesciences
|
|
Keywords:
|
Adaptive test;
survival analysis;
clinical trials
|
|
Abstract:
|
When studying a new population with a clinical trial, there is uncertainty in the underlying assumptions used to estimate the power. This issue is further compounded when a trial has a short enrollment period and a very long follow-up period before the primary endpoint can be analyzed. This can lead to inadequate power for the analysis and the inability to adjust the sample size due to enrollment closing. In the case where two survival curves are to be compared through a fixed point in time, we propose an adaptive test of significance where instead of varying the sample size, the timing for the final analysis is selected between two presepcified time points. This test maximizes the chances of selecting the best time point for the analysis while maintaining the blind and keeping the correct alpha level and power for the test.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2018 program
|
