Motherboard, Reverse Detail: This is a green motherboard, photographed with red-gelled flashes.

The Future of Race-Based Clinical Algorithms

By Jenna Becker

Race-based clinical algorithms are widely used. Yet many race-based adjustments lack evidence and worsen racism in health care. 

Prominent politicians have called for research into the use of race-based algorithms in clinical care as part of a larger effort to understand the public health impacts of structural racism. Physicians and researchers have called for an urgent reconsideration of the use of race in these algorithms. 

Efforts to remove race-based algorithms from practice have thus far been piecemeal. Medical associations, health systems, and policymakers must work in tandem to rapidly identify and remove racist algorithms from clinical practice.

Race-based clinical algorithms

Clinical algorithms are widely used to drive medical care. Some of these algorithms are race-based, “correcting” results based on the race listed in a patient’s medical record. Race-based algorithms can exacerbate existing racial inequalities in health care.

A well-known example of a race-based clinical algorithm is the estimated glomerular filtration rate (eGFR), an equation used to estimate kidney function. This equation is significantly adjusted for Black patients. Some physicians have noted that the eGFR equation disadvantages Black patients by requiring them to become more ill than white patients before they qualify for treatment.

The use of race in clinical algorithms is further complicated by the inaccuracy of racial and ethnic classifications in medical records. Studies have demonstrated that a significant percentage of medical records misclassify patient race. Further, the substantial portion of the population that identifies with multiple racial or ethnic groups is generally not reflected in EHR data or in clinical equations using race. Thus, even if there is a scenario where a race-based equation may be appropriate, it is unclear whether the results would be accurate.

Re-evaluating eGFR equations is a good start, but eGFR is not the only standard that uses race. In fact, a recent study demonstrated that several other clinical algorithms that include race as a factor are in use today. For example, the study noted that the American Heart Association’s Heart Failure Risk score categorized all Black patients as lower risk for heart failure without providing a rationale. This study did not intend to encompass all of medical practice, and therefore likely underestimates the usage of race-based algorithms. 

Due to the widespread usage of race-based clinical algorithms, several politicians have called on the Agency for Healthcare Research and Quality (AHRQ) to research whether race-based equations are warranted. 

Congressional Requests and AHRQ Review

In September, a number of U.S. Congressmembers wrote to AHRQ, requesting that the agency review the use of race-based clinical algorithms in medical practice. The lawmakers sought to understand the extent of the usage of these algorithms and the impact on patient outcomes. This request aligns with AHRQ’s mission to “produce evidence to make health care safer, higher quality, more accessible, equitable, and affordable.”

After the change in administration, AHRQ responded in February with a request for information from clinicians, professional societies, and researchers surrounding the use of race-based algorithms in clinical care. The agency plans to review the race-based algorithms used in clinical practice, the evidence to support their usage, and the potential for bias in these algorithms.

This information will be highly valuable, allowing for the identification and removal of biased medical standards. However, the results of this study may take years to compile and analyze. In the meantime, several groups can work to address this issue.

Moving Forward

Removing racial bias from clinical equations should be a top priority for medical associations, health care organizations, and policymakers. 

As noted in a 2020 New England Journal of Medicine paper, “when clinicians insert race into their tools, they risk interpreting racial disparities as immutable facts rather than as injustices that require intervention.” The authors suggested “reconsidering race correction” to avoid perpetuating racial inequities. Health systems, medical organizations, and policy makers should move forward with such a focus on equity.

It is possible for individual healthcare organizations to stop using race-based equations immediately. For example, several hospital systems have stopped using race-based eGFR equations. Health systems can quickly remove biased race-based equations to avoid perpetuating inequities in various clinical specialties. Health systems can also work to understand and correct the assumptions underlying these algorithms.

Changes can also occur on a specialty level: The National Kidney Foundation and the American Society of Nephrology joined last summer to form a task force to address the race-based equations used for eGFRs. Other organizations can and should follow in that path to review the use of race-based algorithms in their specialty.

Efforts to remove race-based algorithms from practice have not been evenly distributed. The eGFR equations in particular have received significant attention in recent years as an algorithm perpetuating racism. Other equations with a similar potential for bias have not received equal attention. Medical associations and specialty organizations should proactively identify and remove algorithms exacerbating racism in health care. They should not wait for the results of a long review process by AHRQ.

Finally, policymakers should continue to pressure AHRQ and medical associations to prioritize their race-based clinical algorithm research. 

Racist medical algorithms must be addressed urgently. Health care organizations, clinical professional organizations, and policymakers must work in tandem to make that happen.

Jenna Becker

Jenna Becker

Jenna Becker is a 2L at Harvard Law School with a background in healthcare software.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.