Statistics Department Seminar Series: Rob Trangucci, PhD Candidate, Department of Statistics, University of Michigan

Abstract: Despite the importance of vaccine efficacy against post-infection outcomes like transmission or severe illness, these estimands are unidentifiable, even under strong assumptions that are rarely satisfied in real-world trials. We develop a novel method to nonparametrically point identify these principal effects while eliminating the assumptions of monotonicity and perfect infection and post-infection measurements. Furthermore, we show that these results immediately extend to multiple treatments. The result is applicable outside of vaccine efficacy due to the generality of the results. We show that our method can be applied to a variety of clinical trial settings where vaccine efficacy against infection and a post-infection outcome can be jointly inferred. This can yield new insights from existing vaccine efficacy trial data and will aid researchers in designing new multi-arm clinical trials.