Online Program

Return to main conference page
Tuesday, September 24
Tue, Sep 24, 11:45 AM - 1:00 PM
Various Rooms
Roundtable Discussions

TL28: Case Studies in the Use of Real-World Evidence to Improve Regulatory Decision-Making (300859)

*Rebecca A. Hubbard, University of Pennsylvania 
*Brian D. Segal, Flatiron Health 

Keywords: Pragmatic trials, post-marketing surveillance, external control arms

Room: Wilson B

Real-world Evidence (RWE) has the potential to improve regulatory decision making in many ways. Potential uses of RWE include accelerating trials for rare diseases, improving generalizability of clinical studies, serving as external comparators for single-armed trials, and conducting post-marketing safety surveillance. However, real-world data (RWD) have many challenging characteristics not typically encountered in traditional settings. Motivated by the FDA’s RWE framework, the objective of this roundtable is to discuss a series of RWE case-studies, using these to identify challenges and the role of statisticians in overcoming these challenges. For each case study, we will present a description of the study and lead a discussion of the strengths, weaknesses, and challenges of using RWE for this purpose. Case studies will include examples of pragmatic trials conducted within healthcare systems, post-marketing safety surveillance using the FDA Sentinel System, and forming external control arms with the Flatiron Health electronic health record-derived database for comparative effectiveness studies. These case studies will highlight several challenges, including integrating data from clinical records and patient self-report, accounting for heterogeneity across clinical sites, adapting inclusion/exclusion criteria to RWD when key variables are MNAR, handling intermittent missingness in patient data, accounting for irregular assessment schedules, and obtaining reliable endpoints.

We aim to lead a discussion that is informed by our own experience working with RWE as well as elicit the ideas and experiences of roundtable participants. As such, this roundtable will be suitable for statisticians with RWE experience as well as those without experience who want to learn. The goal of the roundtable is for all participants to leave with a practical introduction to RWE, strengths and weaknesses for different use cases, and suggestions for how to mitigate challenges.