A0435
Title: Informed presence in electronic health record data: Bias and bias reduction approaches in longitudinal analyses
Authors: Yun Li - University of Pennsylvania (United States) [presenting]
Abstract: Electronic health record (EHR) systems capture patient information inconsistently, with patients contributing more data when they are sick than when they are healthy. This creates ``informed presence'' and systematic differences between captured and non-captured data biasing estimates of association. There is growing interest in analytic approaches that account for informed presence, but practical approaches for conceptualizing, identifying, and accounting for informed presence in applied EHR-based research have received limited attention. We introduce a collider-bias framework for understanding informed presence bias, novel visualization strategies for irregularly measured data, and four approaches to bias reduction under informed presence. To illustrate, we investigated associations between steroids and cytomegalovirus viremia among pediatric solid organ transplant patients ($N=271$) in a recurrent outcomes analysis. We identified conceptual, descriptive, and analytic evidence of informed presence. Incidence rate ratios dropped from 1.83 (95\% CI: 1.02, 3.28) in a naive analysis to 1.37 (0.73, 2.57) when accounting for informed presence using inverse intensity weighting. When conducting analyses with irregularly measured EHR data, we recommend: 1) identifying the expected observation process using conceptual diagrams; 2) visualizing dependence in the observation process; 3) and accounting for outcome dependence in analyses.
