Virtual
Title: Study Efficacy and Safety of COVID-19 Vaccines using Modern Statistical MethodsAbstract: In this talk, I will present my current research on studying effectiveness and safety of COVID-19 vaccines. In the efficacy studies, we applied the state-of-art statistical techniques to compare BNT162b2, mRNA-1273 and Ad26.COV2.S vaccines against the Delta variant in a large cohort of patients in the Michigan Medicine healthcare system. We also evaluated their effectiveness among individuals who take immunosuppressants. In the safety study, we developed a novel Bayesian graph-assisted signal detection model to mine adverse events of COVID-19 vaccines using VAERS (Vaccine Adverse Event Reporting System). This model allows us to simultaneously estimate all adverse events of vaccines while incorporating ontology of adverse events. Adverse events are naturally related. Explicitly bringing adverse event relationships into the model can aid in the detection of true adverse event signals amid the noise while reducing false positives. Under a fully Bayesian inference framework, we also propose a negative control approach to mitigating the reporting bias and an enrichment approach to detecting AE groups of concern.