Use of race in clinical algorithms

To answer whether patients’ race belongs in clinical prediction algorithms, two types of prediction models are considered: (i) diagnostic, which describes a patient’s clinical characteristics, and (ii) prognostic, which forecasts a clinical risk or treatment effect that a patient is likely to experience in the future. The ex ante equality of opportunity framework is used, where specific health outcomes, which are prediction targets, evolve dynamically due to the effects of legacy levels of outcomes, circumstances, and current individual efforts. In practical settings, this study shows that failure to include race corrections will propagate systemic inequities and discrimination in any diagnostic model and specific prognostic models that inform decisions by invoking an ex ante compensation principle. In contrast, including race in prognostic models that inform resource allocations following an ex ante reward principle can compromise the equality of opportunities for patients from different races. Simulation results demonstrate these arguments.

[1]  C. Manski,et al.  Using measures of race to make clinical predictions: Decision making, patient health, and fairness , 2022, Proceedings of the National Academy of Sciences of the United States of America.

[2]  C. Manski.  Patient-centered appraisal of race-free clinical risk assessment. , 2021, Health Economics.

[3]  N. Powe,et al.  A Unifying Approach for GFR Estimation: Recommendations of the NKF-ASN Task Force on Reassessing the Inclusion of Race in Diagnosing Kidney Disease. , 2021, Journal of the American Society of Nephrology : JASN.

[4]  N. Powe,et al.  A Unifying Approach for GFR Estimation: Recommendations of the NKF-ASN Task Force on Reassessing the Inclusion of Race in Diagnosing Kidney Disease. , 2021, American journal of kidney diseases : the official journal of the National Kidney Foundation.

[5]  Sendhil Mullainathan,et al.  On the Inequity of Predicting A While Hoping for B , 2021 .

[6]  B. Hedden.  On statistical criteria of algorithmic fairness , 2021 .

[7]  R. Neal,et al.  Racial Health Inequities and Clinical Algorithms: A Time for Action. , 2021, Clinical journal of the American Society of Nephrology : CJASN.

[8]  Will Dobbie,et al.  Measuring Racial Discrimination in Algorithms , 2020, AEA Papers and Proceedings.

[9]  David S Jones,et al.  Hidden in Plain Sight - Reconsidering the Use of Race Correction in Clinical Algorithms. , 2020, The New England journal of medicine.

[10]  Jon Kleinberg,et al.  An Economic Perspective on Algorithmic Fairness , 2020 .

[11]  In April,et al.  National Human Genome Research Institute , 2020, Definitions.

[12]  Andrew M. Jones,et al.  Ex ante inequality of opportunity in health, decomposition and distributional analysis of biomarkers. , 2019, Journal of health economics.

[13]  Ravi B. Parikh,et al.  Addressing Bias in Artificial Intelligence in Health Care. , 2019, JAMA.

[14]  Brian W. Powers,et al.  Dissecting racial bias in an algorithm used to manage the health of populations , 2019, Science.

[15]  Andrew M. Jones.  Equity, opportunity and health , 2019, Empirica.

[16]  M. Howell,et al.  Ensuring Fairness in Machine Learning to Advance Health Equity , 2018, Annals of Internal Medicine.

[17]  M. Alsan,et al.  Does Diversity Matter for Health? Experimental Evidence from Oakland , 2018, American Economic Review.

[18]  N. Shah,et al.  Implementing Machine Learning in Health Care - Addressing Ethical Challenges. , 2018, The New England journal of medicine.

[19]  Anthony F. Heath,et al.  Equality of Opportunity , 2017 .

[20]  Alain Trannoy,et al.  Equality of opportunity, moral hazard and the timing of luck , 2017, Soc. Choice Welf..

[21]  Ilya Shpitser,et al.  Fair Inference on Outcomes , 2017, AAAI.

[22]  J. Weinstein,et al.  Communities in Action: Pathways to Health Equity , 2017 .

[23]  Alain Trannoy,et al.  Equality of Opportunity: Theory and Measurement , 2016 .

[24]  Sherri Rose,et al.  A Machine Learning Framework for Plan Payment Risk Adjustment. , 2016, Health services research.

[25]  Alexandra Chouldechova,et al.  Fair prediction with disparate impact: A study of bias in recidivism prediction instruments , 2016, Big Data.

[26]  Jon M. Kleinberg,et al.  Inherent Trade-Offs in the Fair Determination of Risk Scores , 2016, ITCS.

[27]  K. Anderson,et al.  The Private Sector as a Catalyst for Health Equity and a Vibrant Economy: Proceedings of a Workshop , 2016 .

[28]  Cynthia Feliciano.  Shades of Race , 2016 .

[29]  Gary S Collins,et al.  Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD Statement , 2015, BMC Medicine.

[30]  M. Fleurbaey,et al.  Ex Ante Versus Ex Post Equality of Opportunity , 2013 .

[31]  Alain Trannoy,et al.  Equality of opportunity and luck: Definitions and testable conditions, with an application to income in France , 2009 .

[32]  C. Schmid,et al.  A new equation to estimate glomerular filtration rate. , 2009, Annals of internal medicine.

[33]  Mary E. Campbell,et al.  The Implications of Racial Misclassification by Observers , 2007 .

[34]  John E. Roemer,et al.  Equality of opportunity: A progress report , 2002, Soc. Choice Welf..

[35]  T. Mcguire,et al.  Statistical discrimination in health care. , 2001, Journal of Health Economics.

[36]  Larry Alexander,et al.  What Makes Wrongful Discrimination Wrong Biases, Preferences, Stereotypes, and Proxies , 1992 .

[37]  Lawrence B. Joseph.  Some Ways of Thinking About Equality of Opportunity , 1980 .

[38]  J. Charvet.  The Idea of Equality as a Substantive Principle of Society , 1969 .

[39]  Cynthia Feliciano.  Shades of Race: How Phenotype and Observer Characteristics Shape Racial Classification , 2016 .

[40]  Eliminating Racial,et al.  COMMITTEE ON UNDERSTANDING AND ELIMINATING RACIAL AND ETHNIC DISPARITIES IN HEALTH CARE , 2003 .