{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,22]],"date-time":"2026-05-22T23:06:08Z","timestamp":1779491168542,"version":"3.53.1"},"reference-count":35,"publisher":"Oxford University Press (OUP)","issue":"6","license":[{"start":{"date-parts":[[2026,4,6]],"date-time":"2026-04-06T00:00:00Z","timestamp":1775433600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/pages\/standard-publication-reuse-rights"}],"funder":[{"name":"Women\u2019s Auxiliary Millenium Chair"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,6,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>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.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>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.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>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.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>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.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/jamia\/ocag041","type":"journal-article","created":{"date-parts":[[2026,3,17]],"date-time":"2026-03-17T12:34:34Z","timestamp":1773750874000},"page":"1159-1166","source":"Crossref","is-referenced-by-count":0,"title":["Leveraging clinical epidemiology concepts to strengthen machine learning fairness evaluations"],"prefix":"10.1093","volume":"33","author":[{"given":"Lin Lawrence","family":"Guo","sequence":"first","affiliation":[{"name":"Child Health Evaluative Sciences, The Hospital for Sick Children , Toronto, ON M5G1X8,","place":["Canada"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Santiago E","family":"Arciniegas","sequence":"additional","affiliation":[{"name":"Child Health Evaluative Sciences, The Hospital for Sick Children , Toronto, ON M5G1X8,","place":["Canada"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8300-3095","authenticated-orcid":false,"given":"Adam P","family":"Yan","sequence":"additional","affiliation":[{"name":"Child Health Evaluative Sciences, The Hospital for Sick Children , Toronto, ON M5G1X8,","place":["Canada"]},{"name":"Division of Haematology\/Oncology, The Hospital for Sick Children , Toronto, ON M5G1X8,","place":["Canada"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"George A","family":"Tomlinson","sequence":"additional","affiliation":[{"name":"Department of Medicine, University Health Network , Toronto, ON M5G 2C4,","place":["Canada"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4058-1586","authenticated-orcid":false,"given":"Melissa","family":"Beauchemin","sequence":"additional","affiliation":[{"name":"Columbia University School of Nursing\/Herbert Irving Cancer Center , New York, NY 10032,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stephen R","family":"Pfohl","sequence":"additional","affiliation":[{"name":"Google Research, Google , San Francisco, CA 94105,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0951-3091","authenticated-orcid":false,"given":"Lillian","family":"Sung","sequence":"additional","affiliation":[{"name":"Child Health Evaluative Sciences, The Hospital for Sick Children , Toronto, ON M5G1X8,","place":["Canada"]},{"name":"Division of Haematology\/Oncology, The Hospital for Sick Children , Toronto, ON M5G1X8,","place":["Canada"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,4,6]]},"reference":[{"key":"2026052218560896600_ocag041-B1","doi-asserted-by":"crossref","first-page":"100686","DOI":"10.1016\/j.modpat.2024.100686","article-title":"Ethical and bias considerations in artificial intelligence\/machine learning","volume":"38","author":"Hanna","year":"2025","journal-title":"Mod Pathol"},{"key":"2026052218560896600_ocag041-B2","volume-title":"Fairness and Machine Learning: Limitations and Opportunities","author":"Barocas","year":"2023"},{"key":"2026052218560896600_ocag041-B3","year":"1983"},{"key":"2026052218560896600_ocag041-B4","doi-asserted-by":"crossref","first-page":"1462751","DOI":"10.3389\/fdgth.2025.1462751","article-title":"A roadmap to implementing machine learning in healthcare: from concept to practice","volume":"7","author":"Yan","year":"2025","journal-title":"Front Digit Health"},{"issue":"11","key":"2026052218560896600_ocag041-B5","doi-asserted-by":"crossref","first-page":"e40039","DOI":"10.2196\/40039","article-title":"Perspective toward machine learning implementation in pediatric medicine: mixed methods study","volume":"10","author":"Alexander","year":"2022","journal-title":"JMIR Med Inform"},{"key":"2026052218560896600_ocag041-B6","doi-asserted-by":"crossref","DOI":"10.1093\/gigascience\/giac082","article-title":"Statistical quantification of confounding bias in machine learning models","volume":"11","author":"Spisak","year":"2022","journal-title":"Gigascience"},{"key":"2026052218560896600_ocag041-B7","doi-asserted-by":"crossref","first-page":"969","DOI":"10.1093\/ije\/dyad089","article-title":"Co-evolution of epidemiology and artificial intelligence: challenges and opportunities","volume":"52","author":"Sung","year":"2023","journal-title":"Int J Epidemiol"},{"key":"2026052218560896600_ocag041-B8","doi-asserted-by":"crossref","first-page":"17","DOI":"10.1007\/978-1-0716-1138-8_2","volume-title":"Clinical Epidemiology: Practice and Methods","author":"Stuckless","year":"2021"},{"key":"2026052218560896600_ocag041-B9","doi-asserted-by":"crossref","first-page":"e076365","DOI":"10.1136\/bmj-2023-076365","article-title":"Quantifying possible bias in clinical and epidemiological studies with quantitative bias analysis: common approaches and limitations","volume":"385","author":"Brown","year":"2024","journal-title":"BMJ"},{"key":"2026052218560896600_ocag041-B10","first-page":"16","article-title":"Challenges of artificial intelligence in medicine","volume":"323","author":"Aldosari","year":"2025","journal-title":"Stud Health Technol Inform"},{"key":"2026052218560896600_ocag041-B11","doi-asserted-by":"crossref","first-page":"447","DOI":"10.1126\/science.aax2342","article-title":"Dissecting racial bias in an algorithm used to manage the health of populations","volume":"366","author":"Obermeyer","year":"2019","journal-title":"Science"},{"key":"2026052218560896600_ocag041-B12","doi-asserted-by":"crossref","first-page":"1453","DOI":"10.1016\/S0140-6736(17)30569-X","article-title":"Structural racism and health inequities in the USA: evidence and interventions","volume":"389","author":"Bailey","year":"2017","journal-title":"Lancet"},{"key":"2026052218560896600_ocag041-B13","doi-asserted-by":"crossref","first-page":"100864","DOI":"10.1016\/j.patter.2023.100864","article-title":"A normative framework for artificial intelligence as a sociotechnical system in healthcare","volume":"4","author":"McCradden","year":"2023","journal-title":"Patterns (N Y)"},{"key":"2026052218560896600_ocag041-B14","author":"Sendak","year":"2020"},{"issue":"5","key":"2026052218560896600_ocag041-B15","doi-asserted-by":"crossref","first-page":"e384","DOI":"10.1016\/S2589-7500(22)00003-6","article-title":"The medical algorithmic audit","volume":"4","author":"Liu","year":"2022","journal-title":"Lancet Digital Health"},{"key":"2026052218560896600_ocag041-B16","doi-asserted-by":"crossref","first-page":"1124","DOI":"10.1126\/science.185.4157.1124","article-title":"Judgment under uncertainty: heuristics and biases","volume":"185","author":"Tversky","year":"1974","journal-title":"Science"},{"key":"2026052218560896600_ocag041-B17","doi-asserted-by":"crossref","first-page":"230","DOI":"10.1518\/001872097778543886","article-title":"Humans and automation: use, misuse, disuse, abuse","volume":"39","author":"Parasuraman","year":"1997","journal-title":"Hum Factors"},{"key":"2026052218560896600_ocag041-B18","doi-asserted-by":"crossref","first-page":"360","DOI":"10.1038\/s41746-025-01667-2","article-title":"A scoping review and evidence gap analysis of clinical AI fairness","volume":"8","author":"Liu","year":"2025","journal-title":"NPJ Digit Med"},{"key":"2026052218560896600_ocag041-B19","doi-asserted-by":"crossref","first-page":"103621","DOI":"10.1016\/j.jbi.2020.103621","article-title":"An empirical characterization of fair machine learning for clinical risk prediction","volume":"113","author":"Pfohl","year":"2021","journal-title":"J Biomed Inform"},{"issue":"1","key":"2026052218560896600_ocag041-B20","doi-asserted-by":"crossref","first-page":"e100460","DOI":"10.1136\/bmjhci-2021-100460","article-title":"Evaluating algorithmic fairness in the presence of clinical guidelines: the case of atherosclerotic cardiovascular disease risk estimation","volume":"29","author":"Foryciarz","year":"2022","journal-title":"BMJ Health Care Inf"},{"key":"2026052218560896600_ocag041-B21","author":"Liu","year":"2019"},{"key":"2026052218560896600_ocag041-B22","author":"Pfohl","year":"2025"},{"key":"2026052218560896600_ocag041-B23","year":"2026"},{"key":"2026052218560896600_ocag041-B24","year":"2022"},{"key":"2026052218560896600_ocag041-B25","doi-asserted-by":"crossref","first-page":"1667","DOI":"10.1056\/NEJMp068070","article-title":"The challenge of subgroup analyses\u2013reporting without distorting","volume":"354","author":"Lagakos","year":"2006","journal-title":"N Engl J Med"},{"key":"2026052218560896600_ocag041-B26","doi-asserted-by":"crossref","first-page":"2189","DOI":"10.1056\/NEJMsr077003","article-title":"Statistics in medicine-reporting of subgroup analyses in clinical trials","volume":"357","author":"Wang","year":"2007","journal-title":"N Engl J Med"},{"key":"2026052218560896600_ocag041-B27","doi-asserted-by":"crossref","first-page":"101100","DOI":"10.1016\/j.conctc.2023.101100","article-title":"A template for the authoring of statistical analysis plans","volume":"34","author":"Stevens","year":"2023","journal-title":"Contemp Clin Trials Commun"},{"key":"2026052218560896600_ocag041-B28","doi-asserted-by":"crossref","DOI":"10.1136\/bmjhci-2021-100459","article-title":"Conceptualising fairness: three pillars for medical algorithms and health equity","volume":"29","author":"Sikstrom","year":"2022","journal-title":"BMJ Health Care Inform"},{"issue":"1","key":"2026052218560896600_ocag041-B29","doi-asserted-by":"crossref","first-page":"ooae149","DOI":"10.1093\/jamiaopen\/ooae149","article-title":"Algorithmic individual fairness and healthcare: a scoping review","volume":"8","author":"Anderson","year":"2025","journal-title":"JAMIA Open"},{"key":"2026052218560896600_ocag041-B30","doi-asserted-by":"crossref","first-page":"1323","DOI":"10.1093\/jamia\/ocac065","article-title":"A bias evaluation checklist for predictive models and its pilot application for 30-day hospital readmission models","volume":"29","author":"Wang","year":"2022","journal-title":"J Am Med Inform Assoc"},{"key":"2026052218560896600_ocag041-B31","doi-asserted-by":"crossref","first-page":"20230023","DOI":"10.1259\/bjr.20230023","article-title":"AI pitfalls and what not to do: mitigating bias in AI","volume":"96","author":"Gichoya","year":"2023","journal-title":"Br J Radiol"},{"key":"2026052218560896600_ocag041-B32","author":"Dwork","year":"2012"},{"key":"2026052218560896600_ocag041-B33","author":"Kusner","year":"2017"},{"key":"2026052218560896600_ocag041-B34","doi-asserted-by":"crossref","first-page":"473","DOI":"10.1097\/EDE.0000000000000105","article-title":"On the causal interpretation of race in regressions adjusting for confounding and mediating variables","volume":"25","author":"VanderWeele","year":"2014","journal-title":"Epidemiology"},{"key":"2026052218560896600_ocag041-B35","author":"Hanna"}],"container-title":["Journal of the American Medical Informatics Association"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/jamia\/advance-article-pdf\/doi\/10.1093\/jamia\/ocag041\/67871354\/ocag041.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/jamia\/article-pdf\/33\/6\/1159\/67871354\/ocag041.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/jamia\/article-pdf\/33\/6\/1159\/67871354\/ocag041.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,22]],"date-time":"2026-05-22T22:56:15Z","timestamp":1779490575000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jamia\/article\/33\/6\/1159\/8607767"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,6]]},"references-count":35,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2026,4,6]]},"published-print":{"date-parts":[[2026,6,1]]}},"URL":"https:\/\/doi.org\/10.1093\/jamia\/ocag041","relation":{},"ISSN":["1067-5027","1527-974X"],"issn-type":[{"value":"1067-5027","type":"print"},{"value":"1527-974X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,6]]},"published":{"date-parts":[[2026,4,6]]}}}