|
Abstract:
|
To conduct comparative effectiveness research using electronic health records (EHR), many covariates are typically needed to adjust for selection and confounding biases. Unfortunately, it is typical to have missingness in these covariates. Here, we consider the covariates missing at random. Just using cases with complete covariates will result in considerable efficiency losses and likely bias. Standard methods for multiple imputation can either fail to capture nonlinear relationships or suffer from the incompatibility and uncongeniality issues. We explore a flexible Bayesian nonparametric approach to impute the missing covariates which involves factoring the joint distribution of the covariates with missingness into a set of sequential conditionals and applying Bayesian additive regression trees (BART) to model each of these univariate conditionals. We provide details on the computational algorithm and make comparisons to other methods, including parametric sequential imputation and two versions of multiple imputation by chained equations (MICE). We illustrate the proposed approach on EHR data from an affiliated tertiary care institution to examine factors related to hyperglycemia.
|
