|
Activity Number:
|
235
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #310242 |
|
Title:
|
Assessment of the Success of the Propensity Score Estimation and Appropriate Adjustment in a Drug Eluting Stent Post-Marketing Registry
|
|
Author(s):
|
Yun Lu*+ and Kevin Najarian and Aijun Song
|
|
Companies:
|
Boston Scientific Corporation and Boston Scientific Corporation and Boston Scientific Corporation
|
|
Address:
|
100 Boston Scientific Way, Marlborough, MA, 01752,
|
|
Keywords:
|
Propensity Score ; Stepwise Logistic Regression ; Subclassification ; F test ; Drug Eluting Stent
|
|
Abstract:
|
Propensity scores are used frequently in observational studies to adjust for covariate imbalance and reduce bias regarding group comparisons. However, the distribution of propensity scores generated by simple logistic regression may not overlap enough for valid group comparisons. To assess the success of the propensity score estimation and adjustment, covariates, interactions and quadratic terms are included in a logistic regression model with stepwise selection. The distribution of the propensity score is then subclassified into quintiles. F-tests from a sequence of models constructed by gradual refinement of the initial model are used to examine the improvement of the balance across subclasses. Box plots provide visual evaluation of the distributions among groups. This method is applied to a post-marketing drug eluting stent registry study.
|
