10.21.2025 Seminar Series: Challenges in Non-Inferiority Testing: Why It's Not Just a One-Sided Version of Superiority Testing

Seminar Series

Tuesday, October 21, 2025

12:00-1:00PM

Challenges in Non-Inferiority Testing: Why It's Not Just a One-Sided Version of Superiority Testing

Presented by: Dr. Samiran Ghosh, Professor & Chair, Department of Biostatistics & Data Science, UT Health Houston

Abstract: Randomized Controlled Trials (RCTs) are an indispensable source of evidence for evaluating the efficacy of treatments across nearly all disease areas. With the growing availability of multiple treatment options, Comparative Effectiveness Research (CER) has gained increasing importance in supporting informed healthcare decisions. However, the design and analysis of effectiveness trials are significantly more complex than those of traditional efficacy trials. A major factor contributing to this complexity is the inclusion of active comparator arm(s) in RCTs, which gives rise to superiority and non-inferiority (NI) trial designs. Among these, the non-inferiority design plays a fundamental role in CER and will be a key focus of this talk. Non-inferiority (NI) trials are inherently complex, as they typically involve active control and require the specification of a non-inferiority margin a priori. Additional complications arise when the outcome of interest is not continuous, such as binary, count or time-to-event outcomes. In this talk, Dr. Ghosh will provide a broad overview of the key challenges in designing and analyzing NI trials, drawing from his decade-long work in this area. Dr. Ghosh will also present examples highlighting specific difficulties when the outcome of interest is time-to-event and discuss methodological considerations and potential solutions.

Hybrid Seminar: Contact Anirudh Babu (asbabu@utmb.edu) for additional information.