Practical Issues on the Observational Comparative Study Design Using Propensity Score Methodology in Pre-Market Medical Device Clinical Studies from the Regulatory Perspectives

• Activity Number: 297
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Section on Medical Devices and Diagnostics
• Author(s): Nelson Lu*+ and Lilly Yue and Yunling Xu
• Companies: FDA/CDRH and FDA/CDRH and FDA/CDRH
• Keywords: Observational comparative study ; propensity score

Abstract

Observational (non-randomized) comparative studies have often been utilized in the pre-market safety/effectiveness evaluation of therapeutic medical devices due to ethical or practical reasons. The comparators may come from data collected in earlier investigational device exemption studies or registry. To address the possible imbalance in patient characteristics between the investigational device group and the control group, propensity score methodology has been widely used to design and analyze these studies. In this paper, some practical issues and challenges will be discussed from regulatory perspectives. Topics include separation of design and analysis, types of estimands, selection of subjects, sample size estimation, and diagnostic checking in covariate balance.