Encoding Equity
in Clinical Research & Practice
Rethinking race in clinical algorithms

Early life experiences are foundational in shaping health across a lifetime. The consequences of inequitable healthcare in childhood compound over time to exacerbate racial health disparities in adulthood. The American Academy of Pediatrics (AAP) has demonstrated a commitment to eliminating race-based medicine in policy and clinical recommendations to ensure all children have access to affirming, anti-racist health care. The project, “Establishing a Race-Conscious Approach to Clinical Guidance in Pediatric Care,” establishes a systematic process for identifying and correcting bias in clinical guidance and plans to test outcomes based on revised clinical recommendations in a high-volume practice setting.

Eliminating race-based medicine can appear a daunting task as bias, race-correction and/or race-norming are often derived from research that was predicated on race as a biologic construct. Further, seeking the origin of health disparities can lead researchers to cite race, rather than racism, as a driver of disparities. This project demonstrates how to approach the elimination of race-based medicine in clinical guidance, informs future clinical recommendations within the AAP and provides a tested approach for other organizations seeking to rectify the false use of race in clinical guidance. Specifically, the aims of the proposed project include:

• Identify and conduct a rapid revision of three biased, race-corrected and/or race-normed AAP clinical guidelines and develop accompanying guidance to support authors and authoring groups.

• Test the feasibility and effectiveness of implementing an evidence-based strategy to mitigate healthcare inequities in a high-volume acute care clinical setting (e.g., pediatric emergency department).

Learn More

The American Heart Association has designed a two-year strategy to investigate and elevate the complex issue of how race and ethnicity factor into clinical care algorithms and risk prediction tools in cardiovascular science and medicine. The De-biasing Clinical Care Algorithms project represents a targeted infusion of attention and resources within the larger context of the American Heart Association’s commitment to improving health equity and removing structural racism in data science, health research and clinical care delivery in cardiovascular-related domains of medicine. The project has two mutually reinforcing objectives:

• Active engagement of stakeholders and audiences to raise awareness and attention to the use of race in clinical care decision-making tools, including Association staff and volunteers, Association-funded researchers, clinicians and cardiovascular patients/caregivers.

• Contributing novel rigorous evidence to the scientific landscape of race in clinical care algorithms to identify and validate best practices and solutions. The Association has a nearly 100-year track record of curating and advancing rigorous scientific research. Specific awards within the DECCA project complement the larger body of related research annually funded by the Association.

The project leverages significant matched resources with a mix of research funding, expert volunteers, datasets and communications platforms including the Association’s influential scientific conferences and publications.

Learn more:

1 – 2nd round training awards – race in predicting heart disease

2 – Research to test PREVENT risk calculator among diverse groups

3 – 1st round training awards – race in clinical algorithms

The American Medical Association (AMA) co-convenes the Equitable Professional Societies Network (EPSN) alongside the Council of Medical Specialty Societies and in partnership with HealthBegins and Race Forward. The EPSN mobilizes organizations that represent diverse health care professionals; coordinates learning; and encourages action for collective impact so that all people have the power, circumstances, and resources to achieve optimal health. Through coordinated Learning Networks and Action Workstreams, the EPSN equips participating organizations with the knowledge, skills, resources, and tools needed to operationalize equitable structural changes and provides support for applying learning into coordinated actions. Action Workstreams will address topics such as reconsidering race in clinical algorithms and workforce diversification.

Learn more about EPSN here: https://edhub.ama-assn.org/equitable-professional-societies-network

Prior to the launch of the EPSN, the AMA began to track progress on related work by medical societies across the country through its Health Equity in Organized Medicine Survey, with examples published in the Journal of the American Medical Association (JAMA). The AMA also engaged with the Consensus Panel on Racial Bias and Healthcare Algorithms convened by the Agency for Healthcare Research and Quality (AHRQ) and the National Institute on Minority Health and Health Disparities (NIMHD), resulting in the publication of guiding principles to address algorithm bias.

The AMA’s work on harmful race-based algorithms is grounded in its policies on racial essentialism in medicine (D-350.981) and redressing the harms of misusing race in medicine (H-65.943).

The American Medical Student Association (AMSA) is hosting a Doris Duke Foundation Racial Equity in Clinical Equations Civic Science Fellow to develop a more rigorous approach to how race is applied and understood in the design of clinical algorithms and the assessments they inform. The Fellow works to enhance awareness and action through educational programming and advocacy initiatives throughout the organization to influence research practice. 

Learn more

The American Society of Hematology (ASH) is leading a two-year health equity project focused on ensuring that individuals with Duffy-null associated neutrophil count (DANC) receive appropriate care. Two in three people in the U.S. who are of African or Middle Eastern ancestry have the Duffy null phenotype (non-expression of the Duffy antigen on red blood cells).

Of note, race and ethnicity are not biological facts but sociopolitical constructs. Genetic traits do not obey racial boundaries or geographic constructs, so using them as proxies requires extreme caution. While the Duffy null phenotype is not exclusive to African or Arabian Peninsula genetic ancestry, it is most commonly found among these populations.

These individuals are often incorrectly labeled as having neutropenia – a condition that increases the risk of infection or suggests disordered bone marrow function. This mislabeling leads to unnecessary, expensive and invasive testing, delayed or discontinued chemotherapy, exclusion from clinical trials, restricted access to therapeutics and other negative consequences. ASH leads a multi-faceted effort to increase the understanding of Duffy-null associated neutrophil count (DANC) and its impact on the continuum of care. This is accomplished through:

• Empowering selected health care systems to reconsider their ANC ranges for the Duffy-null population and sharing information about their efforts.

• Educating health care professionals about DANC.

• Intervening at the population level, including education of clinical trialists and the public about the implications of DANC.

Learn more

The American Society of Nephrology and the National Kidney Foundation have meticulously reevaluated an algorithm that had used race as a factor in assessing kidney function, a race-based approach that over-estimated Black patients’ kidney function. Among other things, the race-inclusive algorithm inappropriately kept some Black patients off kidney transplant lists. The revised estimated glomerular filtration rate (eGFR) calculator, which now does not include race, has been adopted into practice, and efforts are underway to move up patients on transplant waiting lists who’d erroneously been denied access as a result of the race-based algorithm. ASN and NKF’s partnership was cited by the House Committee on Ways and Means as an example of positive patient-led organizational efforts to address the use of race in clinical decision support tools in a majority report on the issue.

 The American Society of Nephrology and the National Kidney Foundation also championed the removal of race from the Kidney Donor Risk Index, the algorithm that estimates the quality of deceased donor kidneys for transplantation. The race-inclusive algorithm calculates kidneys from Black donors as being of lower quality than kidneys from non-Black donors. These organizational advocacy efforts contributed to the Organ Procurement and Transplant Network (OPTN) to propose a policy change to remove race from the algorithm to better reflect how well the kidney will do when transplanted, which is expected to be finalized.

Learn more

Given concerns about potential harm to Black patients and evidence that a race-neutral approach was superior, an American Thoracic Society (ATS) task force recommended that race and ethnicity no longer be considered factors in interpreting the results of spirometry, the most common type of pulmonary function tests (PFT). Implementation of the race-neutral approach to PFTs is ongoing.

Learn more

Association of Health Care Journalists (AHCJ) is hosting a Doris Duke Foundation Racial Equity in Clinical Equations Civic Science Fellow. The Fellow will focus on increasing awareness of racial bias in health research and how that is communicated in health journalism. Additionally, the Fellow will highlight the patient impact of bias in health research through storytelling and create educational resources for journalists on the issues.

Learn more

The Coalition to End Racism in Clinical Algorithms (CERCA) is a flagship initiative of the NYC Department of Health and Mental Hygiene’s Office of the Chief Medical Officer to address racial inequities in the healthcare delivery system through bridging public health and healthcare in NYC. CERCA’s aims include raising awareness among health system partners on how race adjustment contributes to racial health inequities; elevating and communicating the coalition’s commitment to health equity and race consciousness; measuring institutional impacts of eliminating race adjustment on racial health inequities; and mitigating the harms and delays in care that race adjustment created for patients of color referenced in the algorithms. Due to structural inequities in reimbursement, investment, etc., safety-net health systems – which serve more Medicaid and uninsured people and are located in communities of color, are often under-resourced, and have strained operational capacity. With support from the Doris Duke Foundation, the NYC Health Department provides support  to a group of New York City safety-net health systems in the development and implementation of health equity expert-informed plans to replace at least one of three current algorithms that use race adjustment with a non-race-adjusted algorithm used to assess kidney disease (eGFR), pulmonary disease (PFT) or potential for successful vaginal birth after cesarean section (VBAC).

Learn more

The Council of Medical Specialty Societies (CMSS) has served as co-convener for the 2023 and 2024 convenings on Reconsidering Race and Clinical Algorithms and produced the 2023 report: Driving Equity through New Models in Research and Implementation. CMSS also serves as a host partner for the Doris Duke Racial Equity in Clinical Equations Civic Science Fellows. The CMSS Fellow supports an accelerated and systematic approach to the changes necessary across research, data, clinical guidelines, clinical algorithms, dissemination, and implementation. Working with CMSS and its member societies, they facilitate shared learning across society stakeholders, including society staff and volunteer leaders that oversee equity, clinical guidelines and algorithms, research, informatics, and publishing. This work benefits the nation’s specialty societies, their physician members, and the patients they serve.

Learn more

The National Academies of Sciences, Engineering, and Medicine (NASEM) has convened a multidisciplinary committee to assess the current use of the social constructs of race and ethnicity across the spectrum of biomedical research and provide recommendations to guide the scientific community in the use of race and ethnicity in biomedical research. More specifically, the committee:

• Documents and evaluates how racialized group and ethnic categories are currently being used in biomedical research (e.g., as a descriptor, to stratify data, to apply race norming, to infer differences between groups due to environmental and social impacts), including describing consequences and contributions to health inequities in current clinical practices;
• Identifies the circumstances in which it is appropriate to use the social constructs of race and ethnicity in biomedical research, for example in studying the health effects of racism and the circumstances in which race and ethnicity should not be used to inform inferences;
• Reviews existing guidance for researchers on the use of race as a variable in biomedical research.

Based on its review of the literature and other expert input, the committee is developing a report with its findings, conclusions and recommendations for entities such as researchers, funders, publishers, scientific and medical societies, health systems and industry regarding:

• The use of race and ethnicity in biomedical research, including identifying current practices that should be continued, stopped or modified;
• Policy changes to reform the use of race and ethnicity in biomedical research, with specific attention to the practice of race norming or race correction;
• Implementation strategies to help enhance the adoption of best practices across the biomedical research community.

The committee’s work focuses on the use of racialized group and ethnic categories across the spectrum of biomedical research, including the development of clinical prediction models and other clinical decision tools.
NASEM has also co-sponsored two annual convenings on race and clinical algorithms across medicine, research, and technology.

Learn more: https://www.nationalacademies.org/our-work/the-use-of-race-and-ethnicity-in-biomedical-research

Culture of Health Program Work: https://nam.edu/programs/culture-of-health/systems-impact-special-publication/

Code of Conduct for Artificial Intelligence in Health: The Artificial Intelligence Code of Conduct (AICC) project is an initiative of the NAM Leadership Consortium which aims to: 1) align, share, and broadly support a framework to ensure algorithms in health, health care, and biomedical science perform accurately, safely, reliably, ethically; 2) describe the role of each stakeholder at each stage of the AI lifecycle; 3) describe the national healthcare AI architecture that supports responsible AI; and 4) identify priority action possibilities going forward. This project represents a unique opportunity for national leaders across disciplines to work together to advance trustworthy artificial intelligence in health, healthcare and biomedical science. https://nam.edu/programs/value-science-driven-health-care/health-care-artificial-intelligence-code-of-conduct/

Assessing Meaningful Community Engagement: https://nam.edu/programs/value-science-driven-health-care/assessing-meaningful-community-engagement/

Universities Allied for Essential Medicines (UAEM) is hosting a Doris Duke Foundation Racial Equity in Clinical Equations Civic Science Fellow. The Fellow at UAEM is UAEM’s voice and leader in the racial equity conversation and spearheads efforts to advance the vision around racial equity in clinical equations. The Fellow works alongside the UAEM executive director, board of directors, student volunteers, and organizational partners to deliver a strategy to advance the movement for racial equity in clinical equations. 

Learn more about UAEM

Learn more about UAEM’s Encoding Equity related to Race Equity in Clinical Equations