TY - JOUR
T1 - A bias evaluation checklist for predictive models and its pilot application for 30-day hospital readmission models
AU - Wang, H. Echo
AU - Landers, Matthew
AU - Adams, Roy
AU - Subbaswamy, Adarsh
AU - Kharrazi, Hadi
AU - Gaskin, Darrell J.
AU - Saria, Suchi
N1 - Publisher Copyright:
© 2022 The Author(s). Published by Oxford University Press on behalf of the American Medical Informatics Association.
PY - 2022/8/1
Y1 - 2022/8/1
N2 - Objective: Health care providers increasingly rely upon predictive algorithms when making important treatment decisions, however, evidence indicates that these tools can lead to inequitable outcomes across racial and socio-economic groups. In this study, we introduce a bias evaluation checklist that allows model developers and health care providers a means to systematically appraise a model's potential to introduce bias. Materials and Methods: Our methods include developing a bias evaluation checklist, a scoping literature review to identify 30-day hospital readmission prediction models, and assessing the selected models using the checklist. Results: We selected 4 models for evaluation: LACE, HOSPITAL, Johns Hopkins ACG, and HATRIX. Our assessment identified critical ways in which these algorithms can perpetuate health care inequalities. We found that LACE and HOSPITAL have the greatest potential for introducing bias, Johns Hopkins ACG has the most areas of uncertainty, and HATRIX has the fewest causes for concern. Discussion: Our approach gives model developers and health care providers a practical and systematic method for evaluating bias in predictive models. Traditional bias identification methods do not elucidate sources of bias and are thus insufficient for mitigation efforts. With our checklist, bias can be addressed and eliminated before a model is fully developed or deployed. Conclusion: The potential for algorithms to perpetuate biased outcomes is not isolated to readmission prediction models; rather, we believe our results have implications for predictive models across health care. We offer a systematic method for evaluating potential bias with sufficient flexibility to be utilized across models and applications.
AB - Objective: Health care providers increasingly rely upon predictive algorithms when making important treatment decisions, however, evidence indicates that these tools can lead to inequitable outcomes across racial and socio-economic groups. In this study, we introduce a bias evaluation checklist that allows model developers and health care providers a means to systematically appraise a model's potential to introduce bias. Materials and Methods: Our methods include developing a bias evaluation checklist, a scoping literature review to identify 30-day hospital readmission prediction models, and assessing the selected models using the checklist. Results: We selected 4 models for evaluation: LACE, HOSPITAL, Johns Hopkins ACG, and HATRIX. Our assessment identified critical ways in which these algorithms can perpetuate health care inequalities. We found that LACE and HOSPITAL have the greatest potential for introducing bias, Johns Hopkins ACG has the most areas of uncertainty, and HATRIX has the fewest causes for concern. Discussion: Our approach gives model developers and health care providers a practical and systematic method for evaluating bias in predictive models. Traditional bias identification methods do not elucidate sources of bias and are thus insufficient for mitigation efforts. With our checklist, bias can be addressed and eliminated before a model is fully developed or deployed. Conclusion: The potential for algorithms to perpetuate biased outcomes is not isolated to readmission prediction models; rather, we believe our results have implications for predictive models across health care. We offer a systematic method for evaluating potential bias with sufficient flexibility to be utilized across models and applications.
KW - bias
KW - clinical decision-making
KW - health care disparity
KW - hospital readmission
KW - predictive model
UR - http://www.scopus.com/inward/record.url?scp=85132415857&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85132415857&partnerID=8YFLogxK
U2 - 10.1093/jamia/ocac065
DO - 10.1093/jamia/ocac065
M3 - Article
C2 - 35579328
AN - SCOPUS:85132415857
SN - 1067-5027
VL - 29
SP - 1323
EP - 1333
JO - Journal of the American Medical Informatics Association
JF - Journal of the American Medical Informatics Association
IS - 8
ER -