A0409
Title: A Bayesian nonparametric approach for nonignorable missingness in EHR data
Authors: Sebastien Haneuse - Harvard TH Chan School of Public Health (United States)
David Lindberg - University of Florida (United States)
Michael Daniels - University of Florida (United States) [presenting]
Abstract: An approach for missingness in EHRs is proposed using Bayesian nonparametric (BNP) models. It shows how to introduce sensitivity parameters corresponding to nonignorable missingness in the outcome and confounders by extracting unidentified distributions from the BNP model and reconstructing the distribution of interest. Auxiliary covariates are also flexibly included to move closer to MAR. G-computation is used based on the reconstructed distribution to compute causal estimands of interest.
