Critical Appraisal of Fairness Metrics in Clinical Predictive AI
Not on the map. Why?
Abstract (arXiv)
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.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 45 of its 122 references that are in the map. First posted on arXiv 20 Jun 2025.
Builds on
45 of 122 references.
References · 45
- 2016Equality of Opportunity in Supervised Learningcited by 2,007
- 2017Counterfactual Fairnesscited by 804
- 2014Certifying and removing disparate impactcited by 715
- 2016Fair prediction with disparate impact: A study of bias in recidivism prediction instrumentscited by 672
- 2016Inherent Trade-Offs in the Fair Determination of Risk Scorescited by 586
- 2017Algorithmic decision making and the cost of fairnesscited by 403
- 2020Fairness in Machine Learning: A Surveycited by 370
- 2017Fairness in Criminal Justice Risk Assessments: The State of the Artcited by 315
- 2020Minimax Pareto Fairness: A Multi Objective Perspectivecited by 153
- 2018A Unified Approach to Quantifying Algorithmic Unfairness: Measuring Individual & Group Unfairness via Inequality Indicescited by 97
- 2018The Measure and Mismeasure of Fairnesscited by 95
- 2018Fairness Under Unawareness: Assessing Disparity When Protected Class Is Unobservedcited by 94
33 more
- 2020Why Fairness Cannot Be Automated: Bridging the Gap Between EU Non-Discrimination Law and AIcited by 81
- 2019Putting Fairness Principles into Practice: Challenges, Metrics, and Improvementscited by 70
- 2023Smooth ECE: Principled Reliability Diagrams via Kernel Smoothingcited by 63
- 2023FUTURE-AI: International consensus guideline for trustworthy and deployable artificial intelligence in healthcarecited by 60
- 2024A Closer Look at AUROC and AUPRC under Class Imbalancecited by 57
- 2019The Fairness of Risk Scores Beyond Classification: Bipartite Ranking and the xAUC Metriccited by 41
- 2023A Survey on Intersectional Fairness in Machine Learning: Notions, Mitigation, and Challengescited by 41
- 2023The Unfairness of Fair Machine Learning: Levelling down and strict egalitarianism by defaultcited by 36
- 2023Harvard Glaucoma Fairness: A Retinal Nerve Disease Dataset for Fairness Learning and Fair Identity Normalizationcited by 31
- 2021Soliciting Stakeholders' Fairness Notions in Child Maltreatment Predictive Systemscited by 27
- 2018Creating Fair Models of Atherosclerotic Cardiovascular Disease Riskcited by 23
- 2022Longitudinal Fairness with Censorshipcited by 22
- 2022The four-fifths rule is not disparate impact: a woeful tale of epistemic trespassing in algorithmic fairnesscited by 21
- 2023Improving Fairness in AI Models on Electronic Health Records: The Case for Federated Learning Methodscited by 17
- 2022Net benefit, calibration, threshold selection, and training objectives for algorithmic fairness in healthcarecited by 16
- 2023Coarse race data conceals disparities in clinical risk score performancecited by 15
- 2022Stability of clinical prediction models developed using statistical or machine learning methodscited by 12
- 2020Fair Prediction with Endogenous Behaviorcited by 10
- 2023In the Name of Fairness: Assessing the Bias in Clinical Record De-identificationcited by 9
- 2024A structured regression approach for evaluating model performance across intersectional subgroupscited by 9
- 2024Performance evaluation of predictive AI models to support medical decisions: Overview and guidancecited by 9
- 2022Multi-disciplinary fairness considerations in machine learning for clinical trialscited by 8
- 2023Evaluating the Impact of Social Determinants on Health Prediction in the Intensive Care Unitcited by 7
- 2020Equitable Allocation of Healthcare Resources with Fair Cox Modelscited by 6
- 2023Target specification bias, counterfactual prediction, and algorithmic fairness in healthcarecited by 6
- 2024FairEHR-CLP: Towards Fairness-Aware Clinical Predictions with Contrastive Learning in Multimodal Electronic Health Recordscited by 6
- 2023Bipartite Ranking Fairness through a Model Agnostic Ordering Adjustmentcited by 5
- 2022Detection and Mitigation of Algorithmic Bias via Predictive Rate Paritycited by 4
- 2023(Predictable) Performance Bias in Unsupervised Anomaly Detectioncited by 4
- 2023Fairness-enhancing mixed effects deep learning improves fairness on in- and out-of-distribution clustered (non-iid) datacited by 3
- 2024Evaluating the Impact of Pulse Oximetry Bias in Machine Learning under Counterfactual Thinkingcited by 2
- 2023Effective sample size: a measure of individual uncertainty in predictionscited by 1
- 2024A decomposition of Fisher's information to inform sample size for developing fair and precise clinical prediction models -- part 1: binary outcomescited by 1
Built on by
Citing papers per year: 2026: 2.
2 papers.