Measuring Patient Data Density in EHRs to Understand Bias

The EHR Density Index (EDI), a covariate that quantifies how far a patient's documentation deviates from that of clinically similar patients, developed and validated on approximately 25,000 UNC Health patients as an adjustment for documentation-density bias in EHR studies.

Patients vary widely in how much data their electronic health records contain: sicker patients tend to accumulate more, and existing comorbidity indices like the Charlson and Elixhauser were designed to measure mortality risk; they do not account for documentation density itself. With Emily Pfaff at the NC TraCS Institute, this project develops a covariate that allows analyses to adjust for uneven EHR documentation.

A first manuscript is under review at the Journal of the American Medical Informatics Association. It introduces the EHR Density Index (EDI), which pairs utilization clusters from a Gaussian mixture model with within-cluster residuals across four OMOP domains (conditions, drugs, measurements, and procedures) to characterize how a patient’s documentation deviates from others with similar care patterns. The index was developed and validated on ~25,000 UNC Health patients.