Evaluating Variability with 1:1 Propensity Score-Matched Analyses

Project Title Evaluating Variability with 1:1 Propensity Score-Matched Analyses
Date Posted
Friday, August 18, 2017
Status
In progress
Description

The goal of this project will be to evaluate the sensitivity of results from 1:1 matched analyses related to comparator group sampling. As 1:1 propensity score matching is commonly used in Sentinel Level 2 analyses, it is critical to understand for future assessments using this tool whether the model-based confidence intervals obtained from the Propensity Score tool need to account for additional sources of variability. This project will adopt a plasmode-based simulation framework to examine the impact of accounting for additional sources of uncertainty in PS-matched analyses.

Workgroup Leader(s)

Rishi J. Desai PhD, MS; Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA

Workgroup Members

Younathan Abdia PhD; Jessica Franklin PhD; Joshua J. Gagne PharmD, ScD; Richard Wyss PhD; Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA

Margaret Johnson; Darren Toh ScD; Department of Population Medicine, Harvard Pilgrim Health Care Institute and Harvard Medical School, Boston, MA

Sara Karami PhD, MPH; Center for Drug Evaluation and Research, FDA, Silver Spring, MD 

Hana Lee PhD; Office of Biostatistics, Center for Drug Evaluation and Research, FDA, Silver Spring, MD

Michael Nguyen MD; Office of Surveillance and Epidemiology, Center for Drug Evaluation and Research, FDA, Silver Spring, MD