How to Act Upon Racism—not Race—as a Risk Factor

The present-day practice of trying to avoid consideration of race when making clinical decisions, set next to the historic inclusion of race as a factor in medicine, reveals a tension. Growing evidence of harm has led to new guidance recommending against including race when calculating estimated glomerular filtration rate and in other clinical algorithms.¹ At the same time, federal guidance on risk stratification for COVID-19 suggests considering race or ethnicity, with implications for allocating resources, such as the new oral antiviral treatments and monoclonal antibodies. Why should race be included in one circumstance but not the other? The answer lies in understanding racism—not race—as the risk factor.

The starting point—and what tragically has not changed with time—are extensive racial inequities in health. For generations, as comprehensively described by the Institute of Medicine’s 2002 report, Unequal Treatment, race and ethnicity have been strong predictors of the quality of patient care, even after adjusting for socioeconomic status, age, and other factors. Despite the report’s influence and the many corroborating studies and initiatives^2,3 that have followed it, major inequities in health by race have persisted in the 20 years since the report’s release, with some gains prior to 2020 almost completely reversed by the COVID-19 pandemic.

These inequities were highlighted by the devastating effect COVID-19 has had on Black, Latino, and Indigenous communities throughout the US. The patterns we observed⁴ during New York City’s first spring 2020 wave—with Black and Latino New Yorkers bearing a disproportionate proportion of infections, hospitalizations, and deaths—has continued through the fourth, Omicron variant–driven wave. During this wave, Black New Yorkers with COVID-19 were hospitalized at more than double the rate of White New Yorkers—despite relatively similar vaccination rates. Meanwhile, among patients with positive SARS-CoV-2 test results, Black people and other non-White people, including Asian and Hispanic patients, are documented to have been less likely to receive treatment with monoclonal antibodies,⁵ which could further exacerbate the risk for severe disease.

The discussion on treatment prioritization for COVID-19 must incorporate this baseline reality of unequal treatment for Black, Latino, Asian, Pacific Islander, and Indigenous individuals, despite often greater risk. Even though race is a social construct, exposure to racism has biological consequences. And the goal of using race and ethnicity as a social risk factor is to improve health outcomes of individuals from historically marginalized groups.

Conversely, the effect of using race adjustment in clinical algorithms delays care for Black, Latino, and Indigenous patient populations and upholds a biological definition of race. For example, the “adjustment” factor for the estimated glomerular filtration rate in Black patients calculates kidney function levels to be healthier than in White patients for the same measured result, sometimes delaying necessary care. Similarly, “race norming” in assessing risks in birth outcomes means among persons giving birth, Black or Latino individuals of the same age, and who have the same health status and past birthing history as White individuals, may be significantly more likely to have an unnecessary cesarean delivery. This troubling result must be interpreted in the context of existing birth inequities, particularly higher rates of maternal and infant mortality among Black individuals.

Vyas et al¹ have proposed 3 questions clinicians should ask regarding clinical algorithms: “Is the need for race correction based on robust evidence and statistical analyses? Is there a plausible causal mechanism for the racial difference that justifies the race correction? And would implementing this race correction relieve or exacerbate health inequities?”

The last question could be generalized to the broader issue of when to consider race and ethnicity in practice. The answer is in the outcomes. Given the status quo of pervasive racial inequities, the focus must be on the health effects of using or not using race and ethnicity. In its framework for racial equity impact assessment, the Government Alliance on Race and Equity describes this as improving “overall outcomes by focusing efforts on those who are faring the worst.” Furthermore, this is why more rigorous, complete, and self-identified race and ethnicity data must be collected—across public health and clinical medicine—to ensure that policy decisions are having their intended effect.

What about those who argue that the ends do not justify the means, critiquing the consequentialist philosophical underpinnings of this approach? In this view, prioritizing certain identity groups in treatment allocation or other care leads to a slippery slope of preferential treatment that contravenes values of egalitarianism and fairness. But this view is ahistorical. Approaches that do not consider race, such as allocating COVID-19 treatments based on medical comorbidities alone, have demonstrably failed⁶ to ensure equitable access. In the context of expansive historic and current documentation⁷ of racial and health inequities, considering race and ethnicity is ethically just, in part because it is reparative.⁸ Achieving health equity requires providing resources according to need.

Others may contend that focusing on socioeconomic disadvantage, rather than race, would better serve the aim of health equity. Such a position serves to defang the history of racism, which is foundational, not incidental, to the legacy of the US and its social structures. For example, the pregnancy-related mortality rate for Black women with a college degree or higher is 1.6 times higher than the rate for White women with less than a high school diploma. Experience during COVID-19 has also shown⁹ that adjusting for socioeconomic indicators, such as by using the area deprivation index, fails to fully account for racial and ethnic differences in hospitalizations and other outcomes.

As Jones has described, racism is a system of structuring opportunity and assigning value based on how a person’s appearance is perceived, which unfairly advantages White individuals and communities in the US, unfairly disadvantages other individuals and communities, and saps the strength of the whole of society. Because racism is a race-conscious system, redressing racial inequities requires race-conscious strategies. Adjusting for socioeconomic status may be necessary but is not sufficient because it cannot fully account for the unmeasurable cumulative effects of racism on health and well-being.

Legal counterarguments to addressing racism as a risk factor are also dubious. The authors of the Civil Rights Act of 1964 recognized the need to incorporate race into remedies for racial inequity, ensuring the legality of race-conscious approaches. In New York City, legal authority supports programs like the Minority and Women-owned Business Enterprise Program, which enables businesses that are majority owned, operated, and controlled by individuals who are Black, Hispanic, Asian-Pacific, Asian-Indian, Native American and/or women to both access government funding and gain stronger footholds in fields in which they remain disproportionately underrepresented.

In addition, although there is valid debate about the parameters of race-conscious approaches in health, that debate must be placed into context within a question: What is the alternate approach for redressing racial inequities? If there is agreement that racial inequities are persistent and unacceptable, what is the solution? The onus must be on those arguing against race-conscious strategies for providing that alternative. The status quo for Black, Latino, Asian, Pacific Islander, and Indigenous communities cannot be tolerated.

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2022 Chokshi DA et al. JAMA Health Forum.

Corresponding Author: Dave A. Chokshi, MD, MSc, New York City Department of Health and Mental Hygiene, 42-09 28th St, New York, NY 11101 (dave.chokshi@alumni.duke.edu).

Conflict of Interest Disclosures: None reported.

Additional Contributions: We thank Camara Jones, MD, PhD, MPH, Marc Gourevitch, MD, MPH, and Bram Wispelwey, MD, MPH, for thoughtful feedback on earlier drafts; they were not compensated for their contributions. We also thank Robert Fuller for assistance with an earlier draft.