Algorithms are mathematical formulas and models that combine different input variables or factors to inform a calculation or an estimate—frequently an estimate of risk. In healthcare, algorithms are frequently used to guide clinical decision making both at the point of care and as part of resource allocation and healthcare management. They are increasingly incorporated into healthcare decision tools, such as clinical guidelines, pathways, clinical decision support programs in electronic health records, and operational systems used by health systems and payers.
Although algorithms are widely used and can offer value in diagnostics and treatments, not all individuals benefit equally from such algorithms, creating inequities. This is primarily due to biases that result in undue harm to marginalized populations, such as racial and ethnic minorities, and perpetuate healthcare disparities. Recognition of such disparities has motivated a growing call for clinical algorithms to be both trained and validated on diverse patient data, with representation across spectrums of sex, age, race, ethnicity, and more. To rectify these issues, the field needs to understand when leveraging algorithms leads to unintended biases, how to identify biases before implementation, and what to do with biases discovered after implementation.
In fall 2020, Agency for Healthcare Research and Quality (AHRQ) received a congressional request to review evidence on the use of race and ethnicity within healthcare algorithms, the extent of their use and impact on health disparities, and potential solutions for mitigating racial and ethnic biases to improve disparities and outcomes for racial and ethnic minorities. In addition to commissioning an evidence review through its Evidence-based Practice Center (EPC) Program, AHRQ issued a Request for Information to solicit public input. An analysis of responses has been published.
In March 2023, AHRQ, in partnership with the National Institute on Minority Health and Health Disparities (NIMHD) at the National Institutes of Health, held a 2-day, hybrid meeting to explore the current use of algorithms in healthcare, their impact on racial/ethnic disparities in care, and approaches to identify and mitigate existing biases. AHRQ and NIMHD worked collaboratively with other Federal agencies situated in the Department of Health and Human Services (HHS), including the Office of the National Coordinator for Health Information Technology, the HHS Office of Minority Health, the Centers for Medicare & Medicaid Services, the Office of the Assistant Secretary for Health, and others. Additionally, NORC at the University of Chicago provided support to AHRQ and NIMHD.
The meeting was informed by an evidence report from the EPC Program examining the evidence on algorithms and racial and ethnic bias in healthcare, and approaches to mitigate such bias. Key stakeholders presented additional perspectives on this topic, including initiatives to further explore and address identified challenges.
The meeting was designed to inform the deliberations of a panel of diverse experts representing varied stakeholder perspectives sponsored by AHRQ and NIMHD. The panelists were charged with developing guiding principles and actionable solutions for the use of race and ethnicity within healthcare algorithms during and after the March meeting. A second half-day, virtual meeting was held on May 15 as followup to the March meeting to solicit public feedback on the work of the panel.
The panel’s paper, Guiding Principles to Address the Impact of Algorithm Bias on Racial and Ethnic Disparities in Health and Health Care, was published in JAMA Network Open. The journal also links to an accompanying podcast interview of panel co-chairs Marshall Chin, M.D., M.P.H., and Lucila Ohno-Machado, M.D., Ph.D., M.B.A.
JAMA Network Open citation:
Chin MH, Afsar-Manesh N, Bierman AS, et al. Guiding Principles to Address the Impact of Algorithm Bias on Racial and Ethnic Disparities in Health and Health Care. JAMA Network Open. 2023;6(12):e2345050. DOI: https://doi.org/10.1001/jamanetworkopen.2023.45050.
The evidence review and panel activities will support the field in recognizing the potential for algorithms to mitigate or amplify racial/ethnic bias, understanding how to identify and/or prevent biases before implementation, and understanding how to mitigate biases discovered after implementation.
Meeting materials are below:
- Meeting agenda from March meeting (PDF, 193 KB)
- Slides from March meeting:
- Meeting agenda from May meeting (PDF, 73 KB)
- Results from Request for Information
- Expert Panel Guiding Principles
- AHRQ Evidence report
