A0928
Title: Bayesian methods for vaccine safety surveillance using federated data sources
Authors: Fan Bu - University of Michigan (United States) [presenting]
Abstract: A Bayesian sequential analysis framework for data sources distributed across a federated network motivated by vaccine safety surveillance studies is discussed. The purpose is to enable rapid detection of vaccine safety events from observational healthcare data that accrue over time. Our framework aims at resolving three main challenges: first, control of testing errors in sequential analyses of streaming data; second, correction of bias induced by observational data; third, distributed learning of federated data sources while preserving patient-level privacy. These challenges in a unified statistical framework are tackled by extracting profile likelihoods that retain rich distributional information while protecting individual-level data privacy and hierarchical analysis of adverse control outcomes. As evidenced by large-scale empirical evaluations using real-world data sources, the framework provides substantial improvements over existing approaches to safety surveillance.
