Critical Appraisal of Fairness Metrics in Clinical Predictive AI

TL;DR

This paper critically reviews fairness metrics in clinical predictive AI, highlighting conceptual challenges, gaps in validation, and the need for clinically meaningful measures to ensure equitable healthcare outcomes.

Contribution

It provides a comprehensive classification and critical appraisal of existing fairness metrics in clinical AI, identifying gaps and proposing directions for future research.

Findings

Fragmented landscape of fairness metrics with limited clinical validation

Overreliance on threshold-dependent measures in healthcare

Identified gaps in uncertainty quantification and intersectionality

Abstract

Predictive artificial intelligence (AI) offers an opportunity to improve clinical practice and patient outcomes, but risks perpetuating biases if fairness is inadequately addressed. However, the definition of "fairness" remains unclear. We conducted a scoping review to identify and critically appraise fairness metrics for clinical predictive AI. We defined a "fairness metric" as a measure quantifying whether a model discriminates (societally) against individuals or groups defined by sensitive attributes. We searched five databases (2014-2024), screening 820 records, to include 41 studies, and extracted 62 fairness metrics. Metrics were classified by performance-dependency, model output level, and base performance metric, revealing a fragmented landscape with limited clinical validation and overreliance on threshold-dependent measures. Eighteen metrics were explicitly developed for healthcare, including only one clinical utility metric. Our findings highlight conceptual challenges in defining and quantifying fairness and identify gaps in uncertainty quantification, intersectionality, and real-world applicability. Future work should prioritise clinically meaningful metrics.