aiqbench analysis
This study applies the FairLogue toolkit to clinical machine learning models within the All of Us dataset, demonstrating that intersectional fairness audits uncover significant disparities often missed by single-axis evaluations. Counterfactual analysis reveals that while these compound biases exist in observational data, they largely align with expectations under randomized group membership, suggesting potential confounding factors rather than direct algorithmic discrimination. The work underscores the necessity for rigorous multi-dimensional auditing to ensure equitable deployment of AI systems in healthcare settings.
Evaluating Intersectional Fairness across Clinical Machine Learning Use Cases using Fairlogue and the All of Us Research Program
arXiv:2604.16450v2 Announce Type: replace-cross Abstract: Intersectional biases in healthcare data can produce compound disparities in clinical machine learning models, yet most fairness evaluations assess demographic attributes independently. FairLogue, a toolkit for intersectional fairness auditing, was applied across multiple clinical prediction tasks to evaluate disparities across combined demographic groups. Using the All of Us dataset, two published models were selected for replication and evaluation: (A) prediction of selective serotonin reuptake inhibitor associated bleeding events and (B) two-year stroke risk in patients with atrial fibrillation. Observational fairness metrics were computed across race, gender, and intersectional subgroups, followed by counterfactual analysis to evaluate whether disparities were attributable to group membership. Intersectional evaluation revealed larger disparities than single-axis analyses; however, counterfactual diagnostics indicated that most observed disparities were comparable to those expected under randomized group membership. These results highlight the importance of intersectional fairness auditing and demonstrate how FairLogue provides deeper insight into bias in clinical machine learning systems.
Abstract:Intersectional biases in healthcare data can produce compound disparities in clinical machine learning models, yet most fairness evaluations assess demographic attributes independently. FairLogue, a toolkit for intersectional fairness auditing, was applied across multiple clinical prediction tasks to evaluate disparities across combined demographic groups. Using the All of Us dataset, two published models were selected for replication and evaluation: (A) prediction of selective serotonin reuptake inhibitor associated bleeding events and (B) two-year stroke risk in patients with atrial fibrillation. Observational fairness metrics were computed across race, gender, and intersectional subgroups, followed by counterfactual analysis to evaluate whether disparities were attributable to group membership. Intersectional evaluation revealed larger disparities than single-axis analyses; however, counterfactual diagnostics indicated that most observed disparities were comparable to those expected under randomized group membership. These results highlight the importance of intersectional fairness auditing and demonstrate how FairLogue provides deeper insight into bias in clinical machine learning systems.
Computers and Society (cs.CY); Machine Learning (cs.LG); Quantitative Methods (q-bio.QM)
arXiv:2604.16450 [cs.CY]
(or arXiv:2604.16450v2 [cs.CY] for this version)
https://doi.org/10.48550/arXiv.2604.16450 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 249 words
Related coverage
AI in healthcare: applications and best practices
Databricks · Jul 24, 2026
Beyond Demographic Balance: Multi-Metric and Intersectional Evaluation of Fairness in MIMIC-IV Mortality Prediction
arXiv · Oct 2, 2026
KAISEN: Reproducible Subgroup Fairness Auditing for Clinical Risk Models
arXiv · Jul 31, 2026
VirnyFlow: Optimizing ML Pipelines for Accuracy, Fairness, and Stability at Scale
arXiv · Aug 19, 2026