Javier Cabrera, Rutgers University

Advancing Evidence Generation in Biomedical Research Using Natural Hermite and Propensity Score Indices: Applications to External Control Arms

Methodology for matching distributions has many applications in both clinical and non-clinical study design. The natural Hermite and propensity score indices provide measures of dissimilarity between distributions. These indices, combined with machine learning (AI) algorithms, will be applied for matching distributions in the following applications: • Treatment allocation in animal studies. • The integration of external control arms based on real-world data (RWD) in clinical studies. • Translating the results of a clinical studies to the real world before the release of the treatment. • Adaptive clinical studies that match treatment and control distributions. Simulations results will illustrate the potential gains in power provided by the application of these methods.

BIO: Javier Cabrera has a PhD from Princeton U. Was director of Institute of Biostatistics at Rutgers University and director of Biostatistics at the Cardiovascular Institute of New Jersey. Was chief co-editor of Computational Statistics and Data Analysis, Fulbright fellow, Henry Rutgers fellow, and winner of the SPAIG award from ASA. Supported by grants from NSF, NIH, the RWJ foundation and the Qatar foundation.