Mitigating Health Disparities in EHR via Deconfounder

TL;DR

This paper explores using a deconfounder approach with CVAE to reduce health disparities in EHR predictive models, addressing limitations of traditional fairness methods.

Contribution

It introduces PriMeD, a novel framework employing deconfounder theory and CVAE to mitigate disparities in healthcare datasets.

Findings

Deconfounder can reveal hidden confounders affecting fairness.

PriMeD effectively reduces disparities in experimental results.

Abstract

Health disparities, or inequalities between different patient demographics, are becoming crucial in medical decision-making, especially in Electronic Health Record (EHR) predictive modeling. To ensure the fairness of sensitive attributes, conventional studies mainly adopt calibration or re-weighting methods to balance the performance on among different demographic groups. However, we argue that these methods have some limitations. First, these methods usually mean a trade-off between the model's performance and fairness. Second, many methods completely attribute unfairness to the data collection process, which lacks substantial evidence. In this paper, we provide an empirical study to discover the possibility of using deconfounder to address the disparity issue in healthcare. Our study can be summarized in two parts. The first part is a pilot study demonstrating the exacerbation of disparity when unobserved confounders exist. The second part proposed a novel framework, Parity Medical Deconfounder (PriMeD), to deal with the disparity issue in healthcare datasets. Inspired by the deconfounder theory, PriMeD adopts a Conditional Variational Autoencoder (CVAE) to learn latent factors (substitute confounders) for observational data, and extensive experiments are provided to show its effectiveness.