Modeling Mistrust in End-of-Life Care

TL;DR

This paper develops a machine learning trust score to analyze doctor-patient relationships, revealing racial disparities, the impact of mistrust on outcomes, and differences in patient experiences, advancing fairer healthcare AI.

Contribution

It introduces a trust score derived from machine learning that uncovers disparities and outcomes related to mistrust in end-of-life care.

Findings

Trust score correlates with racial disparities.

Mistrust predicts worse health outcomes.

Patients with higher mistrust report poorer experiences.

Abstract

In this work, we characterize the doctor-patient relationship using a machine learning-derived trust score. We show that this score has statistically significant racial associations, and that by modeling trust directly we find stronger disparities in care than by stratifying on race. We further demonstrate that mistrust is indicative of worse outcomes, but is only weakly associated with physiologically-created severity scores. Finally, we describe sentiment analysis experiments indicating patients with higher levels of mistrust have worse experiences and interactions with their caregivers. This work is a step towards measuring fairer machine learning in the healthcare domain.