Understanding Race Correction in Medical Algorithms: A Critical Examination

Did you know that some tools doctors use to assess your health might be adjusted based on your race? This practice, often called "race correction," is embedded in many clinical algorithms. It is sparking a vital conversation in medicine. While the medical field has debated the role and meaning of race for years, there's still no clear consensus. However, using race correction can perpetuate or even amplify existing health inequities.

What is Race Correction in Medical Algorithms?

Clinical algorithms are tools physicians use to help individualize risk assessments and guide decisions about patient care. These algorithms often consider various patient characteristics. Race correction occurs when the output of these tools is adjusted based on a patient's self-identified race or ethnicity. This practice assumes that race is a reliable proxy for underlying biological or genetic differences. However, mounting evidence suggests this is not the case. Instead, the observed differences often reflect the effects of racism and social factors, which can skew health outcomes.

Examples of Race Correction and Their Impact

Race correction appears in several medical contexts, and its implications can be serious. Here are some specific examples:

Kidney Function (eGFR)

Equations used to estimate glomerular filtration rate (eGFR) from serum creatinine levels often report higher values for patients identified as Black. This adjustment is based on observations of higher average serum creatinine in Black individuals, sometimes attributed to the notion that Black people are more muscular. However, analyses have questioned this claim.

Critics argue that these higher eGFR estimates for Black patients can delay referrals for specialist kidney care or transplantation. Given that Black individuals already experience higher rates of end-stage kidney disease, this practice may worsen health disparities.

Heart Failure Risk

The American Heart Association Get with the Guidelines–Heart Failure Risk Score adds points to the risk score for patients identified as "non-Black." This categorization places Black patients at a perceived lower risk of death. The implications can direct care and resources away from those who need it most.

Cardiac Surgery Risk

Calculators used by the Society of Thoracic Surgeons incorporate race and ethnicity because of observed differences in surgical outcomes. Identifying a patient as "Black/African American" can increase the estimated risk of death by nearly 20% compared to a white patient. Such calculations, when used preoperatively, might steer minority patients, deemed to be at higher risk, away from surgery.

Vaginal Birth After Cesarean (VBAC)

An algorithm predicting the likelihood of a successful VBAC estimates lower chances of success for those identified as African American or Hispanic. Although other factors such as marital status correlated with success, they were not included. Lower estimates of VBAC success for people of color could exacerbate existing disparities, especially since nonwhite U.S. women already have higher rates of cesarean sections, leading to increased maternal mortality rates.

Kidney Donor Risk Index (KDRI)

This index, utilized in kidney allocation, includes the donor's race. Identifying a potential donor as African American raises the predicted risk of kidney graft failure. This assumption could reduce the pool of available kidneys and exacerbate racial inequity in access to kidney transplants.

Osteoporosis Risk (SCORE & FRAX)

Both the SCORE and FRAX tools assess risk for low bone density and fractures. They assign lower risk estimates for Black women compared to white women with identical risk factors. This disparity may delay the necessary evaluation, diagnosis, and treatment for osteoporosis among Black patients.

Pulmonary Function Tests

Spirometers frequently use correction factors that reduce estimated lung function values for individuals labeled as Black (by 10-15%) or Asian (by 4-6%) in the U.S. This can misclassify disease severity and impairment for racial and ethnic minorities in conditions like asthma and COPD.

The Problem with Relying on Correlation

Algorithm developers include race in tools because datasets show a correlation between race/ethnicity and outcomes. However, simply observing a correlation, which may reflect underlying racist social structures, is insufficient. This reliance risks interpreting racial disparities as fixed biological facts. In reality, these disparities stem from racism and social determinants of health that require intervention.

Furthermore, the simplistic categories of race used in these tools do not reflect the complex nature of human diversity and ancestry. This oversimplification complicates decision-making for clinicians trying to apply these adjustments.

Moving Forward: Towards Equitable Algorithms

While race should not be ignored—acknowledging how racism shapes society and health outcomes—its use in prescriptive clinical tools requires critical reevaluation.

Researchers and medical professionals should ask three essential questions when developing or applying algorithms that include race correction:

1. Is the need for race correction based on robust evidence and statistical analyses?

2. Is there a plausible causal mechanism for the racial difference that justifies the correction?

3. Would implementing this race correction relieve or exacerbate health inequities?

   
   

Rigorous analysis is necessary to decode the complex interactions between ancestry, race, racism, socioeconomic status, and the environment. In cases where race adjustments worsen inequities, clinicians and institutions should challenge and revise these tools. Some institutions have already successfully advocated for the removal of race adjustment from tools like eGFR and the VBAC calculator.

A Path to Equity in Healthcare

Ultimately, our clinical tools must evolve to align with our understanding of race and human genetics. By reconsidering race correction, medicine can take a significant step towards becoming a more antiracist field. This change will help ensure that medical practices work to repair, rather than perpetuate, inequities.

Through critical examination and reform of race correction practices, we can strive to create a fairer healthcare landscape that better serves all individuals, promoting health equity and preventing disparities across racial and ethnic lines.

For further discussion on this topic, check out this episode on Spotify.