Evaluating Causal Discovery Algorithms for Path-Specific Fairness and Utility in Healthcare

2026-03-16

AI Summary

This study addresses the challenge of evaluating causal discovery algorithms in medical data, where ground-truth causal graphs are typically unavailable, complicating the joint assessment of fairness and utility. To overcome this, the authors collaborate with domain experts to construct proxy ground-truth causal graphs and evaluate multiple causal discovery algorithms on synthetic clinical datasets for Alzheimer’s disease and heart failure. The work introduces a novel fairness evaluation framework based on path-specific effect decomposition rather than aggregate fairness scores, emphasizing graph-structure-aware fairness analysis. Algorithms including Peter–Clark (PC), Greedy Equivalence Search (GES), and Fast Causal Inference (FCI) are assessed for both structural recovery accuracy and path-specific fairness. Results show that PC achieves the best structural recovery, while FCI yields the highest utility on heart failure data; notably, ejection fraction contributes up to 3.37 percentage points to indirect effects, and significant trade-offs between fairness and utility are observed across algorithms.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Constraint Satisfaction and Optimization: Other Foundations of Constraint Satisfaction

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Responsible Web: Algorithmic accountability and transparency on the web

Abstract

Causal discovery in health data faces evaluation challenges when ground truth is unknown. We address this by collaborating with experts to construct proxy ground-truth graphs, establishing benchmarks for synthetic Alzheimer's disease and heart failure clinical records data. We evaluate the Peter-Clark, Greedy Equivalence Search, and Fast Causal Inference algorithms on structural recovery and path-specific fairness decomposition, going beyond composite fairness scores. On synthetic data, Peter-Clark achieved the best structural recovery. On heart failure data, Fast Causal Inference achieved the highest utility. For path-specific effects, ejection fraction contributed 3.37 percentage points to the indirect effect in the ground truth. These differences drove variations in the fairness-utility ratio across algorithms. Our results highlight the need for graph-aware fairness evaluation and fine-grained path-specific analysis when deploying causal discovery in clinical applications.

Problem

• causal discovery
• path-specific fairness
• healthcare
• ground truth
• fairness-utility tradeoff

Innovation

• causal discovery
• path-specific fairness
• proxy ground-truth graphs
• fairness-utility tradeoff
• clinical causal inference