Write It Like You See It: Detectable Differences in Clinical Notes By Race Lead To Differential Model Recommendations

Abstract

Clinical notes are becoming an increasingly important data source for machine learning (ML) applications in healthcare. Prior research has shown that deploying ML models can perpetuate existing biases against racial minorities, as bias can be implicitly embedded in data. In this study, we investigate the level of implicit race information available to ML models and human experts and the implications of model-detectable differences in clinical notes. Our work makes three key contributions. First, we find that models can identify patient self-reported race from clinical notes even when the notes are stripped of explicit indicators of race. Second, we determine that human experts are not able to accurately predict patient race from the same redacted clinical notes. Finally, we demonstrate the potential harm of this implicit information in a simulation study, and show that models trained on these race-redacted clinical notes can still perpetuate existing biases in clinical treatment decisions.

Cited by in corpus (5)

• Detecting Shortcut Learning for Fair Medical AI using Shortcut Testing
• The Impact and Opportunities of Generative AI in Fact-Checking
• Evaluating the Impact of Social Determinants on Health Prediction in the Intensive Care Unit
• Intersectionality and Testimonial Injustice in Medical Records
• Causal thinking for decision making on Electronic Health Records: why and how