Online Program

Return to main conference page
Friday, September 14
Fri, Sep 14, 9:15 AM - 9:55 AM
Atrium
Poster Session

WITHDRAWN - Use of Propensity Score and Disease Risk Score for Multiple Treatments with Time-to-Event Outcome: A Simulation Study (300672)

Jessica Kim, FDA 
*Di Zhang, University of Pittsburgh 

Keywords: propensity score, disease risk score, time-to-event outcome, multinomial logit model, inverse probability treatment weigh, matching, stratification

Propensity score and disease risk score are often used in pharmacoepidemiologic safety studies. Methods of applying these two balancing scores are extensively studied in binary treatment settings. However, the use of propensity score and disease risk score is not well understood in the case of non-ordinal multiple treatments. Some propensity score methods of multiple treatments have been implemented since the theoretical establishment. Nevertheless, most of the work applies to continuous or binary outcomes. Little work has been done for time-to-event outcomes. In this study, we extend the application of the propensity score and disease risk score methods to time-to-event outcomes in multiple treatment settings. The analytical approaches include weighting, matching, stratification and regression. Simulation studies with rare event rates are conducted to evaluate the performances of different methods. Different treatment-covariates and outcome-covariates strength of associations are considered. Additionally, the impact of large and limited propensity score overlaps is investigated on analytical approaches of propensity scores. The analytical approaches that we recommend in multiple-treatment settings are the inverse probability treatment weighting with bootstrap variance estimator, the generalized propensity score matching, and the Cox regression estimated disease risk score in full cohort. This study aims to provide additional guidance for researchers on propensity score and disease risk score analyses in pharmacoepidemiologic observational studies.