Echoes of Biases: How Stigmatizing Language Affects AI Performance

TL;DR

This study examines how stigmatizing language in electronic health records influences AI performance and racial disparities, revealing that targeted removal of such language by central clinicians can reduce bias effectively.

Contribution

It introduces a novel analysis of clinician biases in EHRs, demonstrating that removing stigmatizing language from central clinicians' notes mitigates racial disparities in AI models.

Findings

Stigmatizing language negatively impacts AI performance for black patients.

Removing SL from central clinicians reduces racial bias more effectively.

SL influences AI outcomes and reflects clinician biases.

Abstract

Electronic health records (EHRs) serve as an essential data source for the envisioned artificial intelligence (AI)-driven transformation in healthcare. However, clinician biases reflected in EHR notes can lead to AI models inheriting and amplifying these biases, perpetuating health disparities. This study investigates the impact of stigmatizing language (SL) in EHR notes on mortality prediction using a Transformer-based deep learning model and explainable AI (XAI) techniques. Our findings demonstrate that SL written by clinicians adversely affects AI performance, particularly so for black patients, highlighting SL as a source of racial disparity in AI model development. To explore an operationally efficient way to mitigate SL's impact, we investigate patterns in the generation of SL through a clinicians' collaborative network, identifying central clinicians as having a stronger impact on racial disparity in the AI model. We find that removing SL written by central clinicians is a more efficient bias reduction strategy than eliminating all SL in the entire corpus of data. This study provides actionable insights for responsible AI development and contributes to understanding clinician behavior and EHR note writing in healthcare.