{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T20:10:41Z","timestamp":1774642241839,"version":"3.50.1"},"reference-count":45,"publisher":"Oxford University Press (OUP)","issue":"11","license":[{"start":{"date-parts":[[2021,8,12]],"date-time":"2021-08-12T00:00:00Z","timestamp":1628726400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000133","name":"Agency for Healthcare Research & Quality","doi-asserted-by":"crossref","award":["K12-HS-026385"],"award-info":[{"award-number":["K12-HS-026385"]}],"id":[{"id":"10.13039\/100000133","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100000026","name":"National Institute for Drug Abuse","doi-asserted-by":"crossref","award":["R25DA035692"],"award-info":[{"award-number":["R25DA035692"]}],"id":[{"id":"10.13039\/100000026","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100000026","name":"National Institute of Drug Abuse","doi-asserted-by":"crossref","award":["R01DA051464"],"award-info":[{"award-number":["R01DA051464"]}],"id":[{"id":"10.13039\/100000026","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100000027","name":"National Institute on Alcohol Abuse and Alcoholism","doi-asserted-by":"publisher","award":["K23AA024503"],"award-info":[{"award-number":["K23AA024503"]}],"id":[{"id":"10.13039\/100000027","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000026","name":"National Institute on Drug Abuse","doi-asserted-by":"publisher","award":["R01DA04171"],"award-info":[{"award-number":["R01DA04171"]}],"id":[{"id":"10.13039\/100000026","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000026","name":"National Institute on Drug Abuse","doi-asserted-by":"publisher","award":["UG1- DA-049467"],"award-info":[{"award-number":["UG1- DA-049467"]}],"id":[{"id":"10.13039\/100000026","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100006108","name":"National Center for Advancing Translational Sciences","doi-asserted-by":"publisher","award":["UL1-TR-002398"],"award-info":[{"award-number":["UL1-TR-002398"]}],"id":[{"id":"10.13039\/100006108","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Dr. Matthew M. Churpeck declares patent pending","award":["ARCD.P0535US.P2"],"award-info":[{"award-number":["ARCD.P0535US.P2"]}]},{"DOI":"10.13039\/100000057","name":"National Institute of General Medical Sciences","doi-asserted-by":"publisher","award":["R01GM123193"],"award-info":[{"award-number":["R01GM123193"]}],"id":[{"id":"10.13039\/100000057","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000092","name":"National Library of Medicine","doi-asserted-by":"publisher","award":["R01LM012973"],"award-info":[{"award-number":["R01LM012973"]}],"id":[{"id":"10.13039\/100000092","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000092","name":"National Library of Medicine","doi-asserted-by":"publisher","award":["R01LM010090"],"award-info":[{"award-number":["R01LM010090"]}],"id":[{"id":"10.13039\/100000092","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2021,10,12]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:sec>\n                  <jats:title>Objectives<\/jats:title>\n                  <jats:p>To assess fairness and bias of a previously validated machine learning opioid misuse classifier.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Materials &amp; Methods<\/jats:title>\n                  <jats:p>Two experiments were conducted with the classifier\u2019s original (n\u2009=\u20091000) and external validation (n\u2009=\u200953 974) datasets from 2 health systems. Bias was assessed via testing for differences in type II error rates across racial\/ethnic subgroups (Black, Hispanic\/Latinx, White, Other) using bootstrapped 95% confidence intervals. A local surrogate model was estimated to interpret the classifier\u2019s predictions by race and averaged globally from the datasets. Subgroup analyses and post-hoc recalibrations were conducted to attempt to mitigate biased metrics.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>We identified bias in the false negative rate (FNR = 0.32) of the Black subgroup compared to the FNR (0.17) of the White subgroup. Top features included \u201cheroin\u201d and \u201csubstance abuse\u201d across subgroups. Post-hoc recalibrations eliminated bias in FNR with minimal changes in other subgroup error metrics. The Black FNR subgroup had higher risk scores for readmission and mortality than the White FNR subgroup, and a higher mortality risk score than the Black true positive subgroup (P\u2009&amp;lt;\u2009.05).<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Discussion<\/jats:title>\n                  <jats:p>The Black FNR subgroup had the greatest severity of disease and risk for poor outcomes. Similar features were present between subgroups for predicting opioid misuse, but inequities were present. Post-hoc mitigation techniques mitigated bias in type II error rate without creating substantial type I error rates. From model design through deployment, bias and data disadvantages should be systematically addressed.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Conclusion<\/jats:title>\n                  <jats:p>Standardized, transparent bias assessments are needed to improve trustworthiness in clinical machine learning models.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1093\/jamia\/ocab148","type":"journal-article","created":{"date-parts":[[2021,7,2]],"date-time":"2021-07-02T03:22:30Z","timestamp":1625196150000},"page":"2393-2403","source":"Crossref","is-referenced-by-count":67,"title":["Bias and fairness assessment of a natural language processing opioid misuse classifier: detection and mitigation of electronic health record data disadvantages across racial subgroups"],"prefix":"10.1093","volume":"28","author":[{"given":"Hale M","family":"Thompson","sequence":"first","affiliation":[{"name":"Department of Psychiatry & Behavioral Sciences, Rush University Medical Center, Chicago, Illinois, USA"}]},{"given":"Brihat","family":"Sharma","sequence":"additional","affiliation":[{"name":"Department of Psychiatry & Behavioral Sciences, Rush University Medical Center, Chicago, Illinois, USA"}]},{"given":"Sameer","family":"Bhalla","sequence":"additional","affiliation":[{"name":"Department of Psychiatry & Behavioral Sciences, Rush University Medical Center, Chicago, Illinois, USA"}]},{"given":"Randy","family":"Boley","sequence":"additional","affiliation":[{"name":"Department of Psychiatry & Behavioral Sciences, Rush University Medical Center, Chicago, Illinois, USA"}]},{"given":"Connor","family":"McCluskey","sequence":"additional","affiliation":[{"name":"Department of Psychiatry & Behavioral Sciences, Rush University Medical Center, Chicago, Illinois, USA"}]},{"given":"Dmitriy","family":"Dligach","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Loyola University, Chicago, Illinois, USA"}]},{"given":"Matthew M","family":"Churpek","sequence":"additional","affiliation":[{"name":"Department of Medicine, University of Wisconsin, Madison, Wisconsin, USA"}]},{"given":"Niranjan S","family":"Karnik","sequence":"additional","affiliation":[{"name":"Department of Psychiatry & Behavioral Sciences, Rush University Medical Center, Chicago, Illinois, USA"}]},{"given":"Majid","family":"Afshar","sequence":"additional","affiliation":[{"name":"Department of Medicine, University of Wisconsin, Madison, Wisconsin, USA"}]}],"member":"286","published-online":{"date-parts":[[2021,8,12]]},"reference":[{"issue":"5","key":"2021101218232339400_ocab148-B1","doi-asserted-by":"crossref","first-page":"878","DOI":"10.1177\/1932296819890623","article-title":"Do-it-yourself automated insulin delivery: a leading example of the democratization of medicine","volume":"14","author":"Burnside","year":"2020","journal-title":"J Diabetes Sci Technol"},{"issue":"7","key":"2021101218232339400_ocab148-B2","doi-asserted-by":"crossref","first-page":"961","DOI":"10.1016\/j.jacr.2019.04.023","article-title":"Democratizing AI","volume":"16","author":"Allen","year":"2019","journal-title":"J Am Coll Radiol"},{"key":"2021101218232339400_ocab148-B3","author":"Gupta","year":"2020"},{"issue":"1","key":"2021101218232339400_ocab148-B4","doi-asserted-by":"crossref","first-page":"79","DOI":"10.1186\/s12911-020-1099-y","article-title":"Publicly available machine learning models for identifying opioid misuse from the clinical notes of hospitalized patients","volume":"20","author":"Sharma","year":"2020","journal-title":"BMC Med Inform Decis Mak"},{"issue":"3","key":"2021101218232339400_ocab148-B5","doi-asserted-by":"crossref","first-page":"254","DOI":"10.1093\/jamia\/ocy166","article-title":"Natural language processing and machine learning to identify alcohol misuse from the electronic health record in trauma patients: development and internal validation","volume":"26","author":"Afshar","year":"2019","journal-title":"J Am Med Inform Assoc"},{"issue":"7","key":"2021101218232339400_ocab148-B6","doi-asserted-by":"crossref","first-page":"e0219717","DOI":"10.1371\/journal.pone.0219717","article-title":"Subtypes in patients with opioid misuse: a prognostic enrichment strategy using electronic health record data in hospitalized patients","volume":"14","author":"Afshar","year":"2019","journal-title":"PLoS One"},{"key":"2021101218232339400_ocab148-B7","first-page":"587","article-title":"Data statements for natural language processing: toward mitigating system bias and enabling better science","volume":"6","author":"Bender","year":"2018","journal-title":"Trans Asscoc Comput Linguist"},{"key":"2021101218232339400_ocab148-B8","first-page":"289","author":"Benthall","year":"2019"},{"issue":"6464","key":"2021101218232339400_ocab148-B9","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":"2021101218232339400_ocab148-B10","first-page":"67","author":"Dixon","year":"2018"},{"issue":"11","key":"2021101218232339400_ocab148-B11","doi-asserted-by":"crossref","first-page":"35","DOI":"10.1080\/15265161.2020.1820102","article-title":"Structural disparities in data science: a prolegomenon for the future of machine learning","volume":"20","author":"Karnik","year":"2020","journal-title":"Am J Bioeth"},{"key":"2021101218232339400_ocab148-B12","doi-asserted-by":"crossref","first-page":"183","DOI":"10.1016\/j.drugpo.2019.01.010","article-title":"The triple wave epidemic: supply and demand drivers of the US opioid overdose crisis","volume":"71","author":"Ciccarone","year":"2019","journal-title":"Int J Drug Policy"},{"key":"2021101218232339400_ocab148-B13","doi-asserted-by":"crossref","first-page":"112","DOI":"10.1016\/j.drugpo.2017.06.003","article-title":"US regional and demographic differences in prescription opioid and heroin-related overdose hospitalizations","volume":"46","author":"Unick","year":"2017","journal-title":"Int J Drug Policy"},{"issue":"2","key":"2021101218232339400_ocab148-B14","doi-asserted-by":"crossref","first-page":"217","DOI":"10.1057\/biosoc.2015.46","article-title":"White opioids: pharmaceutical race and the war on drugs that wasn\u2019t","volume":"12","author":"Netherland","year":"2017","journal-title":"Biosocieties"},{"issue":"12","key":"2021101218232339400_ocab148-B15","doi-asserted-by":"crossref","first-page":"2127","DOI":"10.2105\/AJPH.2016.303483","article-title":"Is the prescription opioid epidemic a white problem?","volume":"106","author":"Hansen","year":"2016","journal-title":"Am J Public Health"},{"issue":"4","key":"2021101218232339400_ocab148-B16","doi-asserted-by":"crossref","first-page":"398","DOI":"10.1080\/09581596.2019.1626002","article-title":"Intersectional inequalities and the US opioid crisis: challenging dominant narratives and revealing heterogeneities","volume":"30","author":"Persmark","year":"2020","journal-title":"Crit Public Health"},{"issue":"8","key":"2021101218232339400_ocab148-B17","doi-asserted-by":"crossref","first-page":"1127","DOI":"10.1007\/s10995-010-0720-6","article-title":"Universal alcohol\/drug screening in prenatal care: a strategy for reducing racial disparities? Questioning the assumptions","volume":"15","author":"Roberts","year":"2011","journal-title":"Matern Child Health J"},{"issue":"4","key":"2021101218232339400_ocab148-B18","doi-asserted-by":"crossref","first-page":"e203711","DOI":"10.1001\/jamanetworkopen.2020.3711","article-title":"Association of racial\/ethnic segregation with treatment capacity for opioid use disorder in counties in the United States","volume":"3","author":"Goedel","year":"2020","journal-title":"JAMA Netw Open"},{"issue":"1","key":"2021101218232339400_ocab148-B19","doi-asserted-by":"crossref","first-page":"98","DOI":"10.2105\/AJPH.2020.306006","article-title":"Artificial intelligence, intersectionality, and the future of public health","volume":"111","author":"Bauer","year":"2021","journal-title":"Am J Public Health"},{"key":"2021101218232339400_ocab148-B20","author":"Saleiro","year":"2018"},{"key":"2021101218232339400_ocab148-B21","first-page":"1135","author":"Ribeiro","year":"2016"},{"issue":"6","key":"2021101218232339400_ocab148-B22","doi-asserted-by":"crossref","first-page":"460","DOI":"10.1097\/ADM.0000000000000520","article-title":"The substance use intervention team: a preliminary analysis of a population-level strategy to address the opioid crisis at an Academic Health Center","volume":"13","author":"Thompson","year":"2019","journal-title":"J Addict Med"},{"issue":"10","key":"2021101218232339400_ocab148-B23","doi-asserted-by":"crossref","first-page":"e0239761","DOI":"10.1371\/journal.pone.0239761","article-title":"Differences in length of stay and discharge destination among patients with substance use disorders: the effect of Substance Use Intervention Team (SUIT) consultation service","volume":"15","author":"Thompson","year":"2020","journal-title":"PLoS One"},{"issue":"2","key":"2021101218232339400_ocab148-B24","doi-asserted-by":"crossref","first-page":"189","DOI":"10.1016\/j.jsat.2006.08.002","article-title":"A comprehensive review of the psychometric properties of the Drug Abuse Screening Test","volume":"32","author":"Yudko","year":"2007","journal-title":"J Subst Abuse Treat"},{"issue":"1","key":"2021101218232339400_ocab148-B25","doi-asserted-by":"crossref","first-page":"19","DOI":"10.1186\/s13722-021-00229-7","article-title":"External validation of an opioid misuse machine learning classifier in hospitalized adult patients","volume":"16","author":"Afshar","year":"2021","journal-title":"Addict Sci Clin Pract"},{"key":"2021101218232339400_ocab148-B26","year":"2019"},{"issue":"35","key":"2021101218232339400_ocab148-B27","doi-asserted-by":"crossref","first-page":"1189","DOI":"10.15585\/mmwr.mm6935a1","article-title":"Vital signs: characteristics of drug overdose deaths involving opioids and stimulants - 24 states and the district of Columbia, January-June 2019","volume":"69","author":"O'Donnell","year":"2020","journal-title":"MMWR Morb Mortal Wkly Rep"},{"issue":"4","key":"2021101218232339400_ocab148-B28","doi-asserted-by":"crossref","first-page":"e0214906","DOI":"10.1371\/journal.pone.0214906","article-title":"Alcohol use, acculturation and socioeconomic status among Hispanic\/Latino men and women: The Hispanic Community Health Study\/Study of Latinos","volume":"14","author":"Casta\u00f1eda","year":"2019","journal-title":"PLoS One"},{"key":"2021101218232339400_ocab148-B29","doi-asserted-by":"crossref","first-page":"61","DOI":"10.1016\/j.annepidem.2020.06.007","article-title":"A longitudinal analysis of nondaily smokers: the Hispanic Community Health Study\/Study of Latinos (HCHS\/SOL)","volume":"49","author":"Klugman","year":"2020","journal-title":"Ann Epidemiol"},{"key":"2021101218232339400_ocab148-B30","author":"Ribeiro"},{"key":"2021101218232339400_ocab148-B31","volume-title":"Interpretable Machine Learning: A Guide for Making Black Box models Explainable","author":"Molnar"},{"issue":"1","key":"2021101218232339400_ocab148-B32","doi-asserted-by":"crossref","first-page":"73","DOI":"10.1097\/01.ede.0000147512.81966.ba","article-title":"Optimal cut-point and its corresponding Youden Index to discriminate individuals using pooled blood samples","volume":"16","author":"Schisterman","year":"2005","journal-title":"Epidemiology"},{"issue":"4","key":"2021101218232339400_ocab148-B33","doi-asserted-by":"crossref","first-page":"621","DOI":"10.1093\/jamia\/ocz228","article-title":"A tutorial on calibration measurements and calibration models for clinical prediction models","volume":"27","author":"Huang","year":"2020","journal-title":"J Am Med Inform Assoc"},{"issue":"6","key":"2021101218232339400_ocab148-B34","doi-asserted-by":"crossref","first-page":"626","DOI":"10.1097\/MLR.0b013e31819432e5","article-title":"A modification of the Elixhauser comorbidity measures into a point system for hospital death using administrative data","volume":"47","author":"van Walraven","year":"2009","journal-title":"Med Care"},{"key":"2021101218232339400_ocab148-B35","volume-title":"Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care","author":"Smedley","year":"2003"},{"key":"2021101218232339400_ocab148-B36","year":"2020"},{"issue":"16","key":"2021101218232339400_ocab148-B37","doi-asserted-by":"crossref","first-page":"4296","DOI":"10.1073\/pnas.1516047113","article-title":"Racial bias in pain assessment and treatment recommendations, and false beliefs about biological differences between blacks and whites","volume":"113","author":"Hoffman","year":"2016","journal-title":"Proc Natl Acad Sci USA"},{"issue":"22","key":"2021101218232339400_ocab148-B38","doi-asserted-by":"crossref","first-page":"3152","DOI":"10.1093\/bioinformatics\/btu520","article-title":"Bias correction for selecting the minimal-error classifier from many machine learning models","volume":"30","author":"Ding","year":"2014","journal-title":"Bioinformatics"},{"issue":"4","key":"2021101218232339400_ocab148-B39","doi-asserted-by":"crossref","first-page":"e213909","DOI":"10.1001\/jamanetworkopen.2021.3909","article-title":"Comparison of methods to reduce bias from clinical prediction models of postpartum depression","volume":"4","author":"Park","year":"2021","journal-title":"JAMA Netw Open"},{"key":"2021101218232339400_ocab148-B40","doi-asserted-by":"crossref","first-page":"i6","DOI":"10.1136\/bmj.i6","article-title":"Net benefit approaches to the evaluation of prediction models, molecular markers, and diagnostic tests","volume":"352","author":"Vickers","year":"2016","journal-title":"BMJ"},{"issue":"1","key":"2021101218232339400_ocab148-B41","doi-asserted-by":"crossref","first-page":"5131","DOI":"10.1038\/s41467-020-18918-3","article-title":"Deep transfer learning for reducing health care disparities arising from biomedical data inequality","volume":"11","author":"Gao","year":"2020","journal-title":"Nat Commun"},{"issue":"12","key":"2021101218232339400_ocab148-B42","doi-asserted-by":"crossref","first-page":"866","DOI":"10.7326\/M18-1990","article-title":"Ensuring fairness in machine learning to advance health equity","volume":"169","author":"Rajkomar","year":"2018","journal-title":"Ann Intern Med"},{"key":"2021101218232339400_ocab148-B43","first-page":"1989: 8","article-title":"Demarginalizing the Intersection of Race and Sex: A Black Feminist Critique of Antidiscrimination Doctrine, Feminist Theory and Antiracist Politics","author":"Crenshaw","year":"1989","journal-title":"Univ Chic Leg Forum"},{"issue":"6","key":"2021101218232339400_ocab148-B44","doi-asserted-by":"crossref","first-page":"1241","DOI":"10.2307\/1229039","article-title":"Mapping the margins: intersectionality, identity politics, and violence against women of color","volume":"43","author":"Crenshaw","year":"1991","journal-title":"Stanford Law Rev"},{"issue":"7","key":"2021101218232339400_ocab148-B45","doi-asserted-by":"crossref","first-page":"1267","DOI":"10.2105\/AJPH.2012.300750","article-title":"The problem with the phrase women and minorities: intersectionality-an important theoretical framework for public health","volume":"102","author":"Bowleg","year":"2012","journal-title":"Am J Public Health"}],"container-title":["Journal of the American Medical Informatics Association"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/academic.oup.com\/jamia\/article-pdf\/28\/11\/2393\/40576428\/ocab148.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"http:\/\/academic.oup.com\/jamia\/article-pdf\/28\/11\/2393\/40576428\/ocab148.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,10,17]],"date-time":"2021-10-17T22:41:25Z","timestamp":1634510485000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jamia\/article\/28\/11\/2393\/6349190"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,8,12]]},"references-count":45,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2021,8,12]]},"published-print":{"date-parts":[[2021,10,12]]}},"URL":"https:\/\/doi.org\/10.1093\/jamia\/ocab148","relation":{},"ISSN":["1527-974X"],"issn-type":[{"value":"1527-974X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2021,11,1]]},"published":{"date-parts":[[2021,8,12]]}}}