Bias in Real-world Endpoints when Conducting External Control Arm Analyses

When augmenting single-arm trials with real-world data for external control arm analyses, biases may arise due to differences in how and when patients’ disease progression is assessed between the two settings. Funded by an FDA U01 methods research grant, I’ve developed methods and simulation tools to quantify the impact of these biases, calibrate real-world endpoints to be more trial-like, and conduct Quantitative Bias Analyses to contextualize ECA findings in the presence of endpoint measurement error.

Extending Inferences from Trials to Target Populations

Patients recruited for clinical trials are often not representative of the target populations of interest. I’ve developed methods to assess and improve upon the relevance of trial findings to target populations, transporting findings from trial data using population reference databases (i.e. large health surveys, electronic health records, etc.)