Leveraging clinical epidemiology concepts to strengthen machine learning fairness evaluations
Journal of the American Medical Informatics Association · 01.06.2026 · 33 (6) · 1159-1166
DOI: 10.1093/jamia/ocag041 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of the American Medical Informatics Association, online erschienen am 06.04.2026. Verfasst von Lin Lawrence Guo und 6 weiteren Autor:innen. Beteiligte Einrichtung: Child Health Evaluative Sciences, The Hospital for Sick Children, Toronto, ON M5G1X8, Canada. Hauptthemen laut MEDLINE: Machine Learning. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: Women's Auxiliary Millenium Chair supported the PREDICT, Women's Auxiliary Millenium Chair, Women’s Auxiliary Millenium Chair.
Abstract (Originalsprache)
Objectives
The increasing use of machine learning (ML) in clinical care makes fairness a central issue. Fairness, defined as the absence of disparities across individuals or subgroups, shares several parallels with concepts in clinical epidemiology. The objective was to apply clinical epidemiology frameworks to the evaluation of ML fairness, both to enhance understanding of these concepts and to strengthen fairness assessments.
Methods
This manuscript addresses: (1) the connection between clinical epidemiology bias terms and ML fairness concepts; (2) the relationship between diagnostic testing metrics and fairness criteria; (3) issues arising from multiple testing; and (4) strategies for fairness considerations leveraging clinical epidemiology principles.
Results
Unfairness can arise at different stages: before model development, during model development and post-deployment. The root of unfairness can be conceptualized as a result of clinical epidemiology bias concepts, such as selection, measurement, model misspecification, cognitive and implementation bias. Common approaches to evaluating fairness involve comparing model performance metrics across 1 or more subgroups. Four widely used fairness criteria are independence, separation, sufficiency and predictive parity. They can be assessed using diagnostic testing metrics. Fairness evaluations are vulnerable to multiple testing issues, with subgroup analyses posing particular risks for spurious findings. Solutions can leverage established clinical epidemiology principles such as pre-specifying the analytic strategy.
Conclusions
Many parallels exist between ML fairness and clinical epidemiology, including the conceptualization of the root causes of unfairness, the articulation of fairness criteria, and considerations related to multiple testing. Methodologically sound fairness approaches can leverage well-established principles from clinical epidemiology.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Child Health Evaluative Sciences, The Hospital for Sick Children, Toronto, ON M5G1X8, Canada.
- Division of Haematology/Oncology, The Hospital for Sick Children, Toronto, ON M5G1X8, Canada.
- Department of Medicine, University Health Network, Toronto, ON M5G 2C4, Canada.
- Columbia University School of Nursing/Herbert Irving Cancer Center, New York, NY 10032, United States.
- Google Research, Google, San Francisco, CA 94105, United States.
ORCID-Profile: Adam P Yan, Melissa Beauchemin, Lillian Sung
Förderung
- Women's Auxiliary Millenium Chair supported the PREDICT
- Women's Auxiliary Millenium Chair
- Women’s Auxiliary Millenium Chair
Zitieren
APA
Guo, L. L., Arciniegas, S. E., Yan, A. P., Tomlinson, G. A., Beauchemin, M., Pfohl, S. R., & Sung, L. (2026). Leveraging clinical epidemiology concepts to strengthen machine learning fairness evaluations. Journal of the American Medical Informatics Association, 33(6), 1159-1166. https://doi.org/10.1093/jamia/ocag041
Vancouver
Guo LL, Arciniegas SE, Yan AP, Tomlinson GA, Beauchemin M, Pfohl SR, et al. Leveraging clinical epidemiology concepts to strengthen machine learning fairness evaluations. Journal of the American Medical Informatics Association. 2026;33(6):1159-1166. doi:10.1093/jamia/ocag041
BibTeX
@article{Guo2026,
author = {Guo, Lin Lawrence and Arciniegas, Santiago E and Yan, Adam P and Tomlinson, George A and Beauchemin, Melissa and Pfohl, Stephen R and Sung, Lillian},
title = {Leveraging clinical epidemiology concepts to strengthen machine learning fairness evaluations},
journal = {Journal of the American Medical Informatics Association},
year = {2026},
volume = {33},
number = {6},
pages = {1159--1166},
doi = {10.1093/jamia/ocag041}
}
EUVIMED