|
Activity Number:
|
495
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Thursday, August 2, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #309629 |
|
Title:
|
Assessing Correlation Between Repeated Measurements of Secondary Endpoints in a Medical Device Trial
|
|
Author(s):
|
Lan Pan*+ and Bryan Randall and Alex Shih
|
|
Companies:
|
Boston Scientific Corporation and Boston Scientific Corporation and Boston Scientific Corporation
|
|
Address:
|
One SCIMED Place A242, Maple Grove, MN, 55311,
|
|
Keywords:
|
Longitudinal Methods ; General Linear Model ; Structured Covariance ; Missing Values
|
|
Abstract:
|
Change in hypertension control is one of the important secondary endpoints in a Boston Scientific Renal Stent System trial. Blood pressures were measured several times during the early and long-term follow-up. In previous analyses of the trial data, only the cross-sectional measurements of blood pressure were used to investigate the change in hypertension control between two baseline resistive index subgroups. In this presentation, we conduct the longitudinal data analysis to assess the correlation between repeated measurements and provide more efficient estimators. We conduct multiple model comparisons to select the most appropriate covariance structure. We also fit the model on balanced and unbalanced data to examine the impact of missing values. The results do not show baseline resistive index as a predictor of blood pressure change in this patient population.
|
