Nima Hejazi, currently a Postdoctoral Research Fellow in Biostatistics in the Department of Population Health Sciences at Weill Cornell Medicine and a candidate for a faculty position in the Division of Biostatistics, will present:

“Efficiently Evaluating the Causal Impacts of Continuous Exposures in Vaccine Trials”

Abstract: Continuous exposures pose significant challenges for causal inference, yet such variables abound in scientific practice. Static interventions, which set the exposure to a fixed, pre-specified value across all units, yield causal effects that are challenging to identify and estimate. Stochastic interventions instead define counterfactuals via hypothetical shifts of the exposure distribution, a model-free generalization of the slope of a linear model. We present novel non/semi-parametric efficient estimators of these effects, using frameworks that incorporate state-of-the-art machine learning. We illustrate how our techniques can be used to delineate the causal impacts of vaccine-induced immune responses on infection/disease endpoints in applications to the HIV Vaccine Trial Network’s 505 HIV-1 and the US Government / Coronavirus Vaccine Prevention Network’s COVE (Moderna) COVID-19 vaccine trials.

All are welcome.