Causal Inference for Estimation of Vaccine Effects from Time-to-Event Data

Abstract

AbstractVaccine is the most efficient method for controlling of infectious disease. Vaccine effectiveness estimation is extremely important in monitoring vaccine efficacy and controlling disease spreading. To study about the COVID-19 vaccine effectiveness from EHR data, we apply the counterfactual reasoning method with deep neural network for vaccine effectiveness estimation from the time-to-event data which are extracted from Optum EHR dataset. The estimated vaccine effectiveness by the counterfactual reasoning is compared with the Cox regression model and Random survival forest model. The preliminary results show that the proposed model is more unbiased than the Cox regression and Random survival forest models.