Examining Imbalance Effects on Performance and Demographic Fairness of Clinical Language Models

2024-12-23

arXiv.org

AI Summary

This study investigates how data imbalance affects both predictive performance and demographic fairness—across gender, age, race, and social determinants of health—in clinical language models for ICD coding. We propose a unified “imbalance–performance–fairness” triadic analytical framework, integrating multidimensional imbalance metrics, subgroup fairness evaluations (e.g., TPR difference, equalized odds gap), and causal inference techniques, applied systematically to state-of-the-art biomedical language models. Through rigorous statistical attribution analysis, we make the novel finding that intra-majority-class feature homogeneity—not merely class-level sample imbalance—is the primary driver of fairness disparities. This insight shifts the diagnostic focus in clinical NLP from aggregate class distribution toward fine-grained representation geometry within majority groups, offering actionable levers for fairness-aware model design and intervention.

Technology Category

• Machine Learning: Ethics, Bias, and Fairness
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy
• Philosophy and Ethics of AI: Bias, Fairness & Equity

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Data imbalance is a fundamental challenge in applying language models to biomedical applications, particularly in ICD code prediction tasks where label and demographic distributions are uneven. While state-of-the-art language models have been increasingly adopted in biomedical tasks, few studies have systematically examined how data imbalance affects model performance and fairness across demographic groups. This study fills the gap by statistically probing the relationship between data imbalance and model performance in ICD code prediction. We analyze imbalances in a standard benchmark data across gender, age, ethnicity, and social determinants of health by state-of-the-art biomedical language models. By deploying diverse performance metrics and statistical analyses, we explore the influence of data imbalance on performance variations and demographic fairness. Our study shows that data imbalance significantly impacts model performance and fairness, but feature similarity to the majority class may be a more critical factor. We believe this study provides valuable insights for developing more equitable and robust language models in healthcare applications.

Problem

• Impact of data imbalance on model performance
• Influence of data imbalance on demographic fairness
• Feature similarity effects in ICD code prediction

Innovation

• Analyzing data imbalance effects
• Exploring demographic fairness impacts
• Deploying diverse performance metrics