Racial biases in medical decision-making tools

Perceived differences in race were used to justify slavery in the United States and these beliefs impacted medical treatment and experimentation on Black persons in the United States.[8] Medical treatment of Black persons in the United States came into specific focus after 1808, after a federal ban on slave imports was implemented.[9] Physicians such as J Marion Sims started to focus on the existing Black population, attempting to observe and use Black slaves for medical reasons.[10][9] Some of these treatments and experiments were for the broader medical community, others were to increase birth rates and therefore domestic slave populations. Sims, for instance, the father of modern gynecology, performed surgery on twelve women in the 1840s from his backyard in Montgomery, Alabama.[11] All of the women were enslaved.[11] As modern medical science grew, it developed in conjunction with notions of racism, derived from both experiments and societal sentiments. These factors embedded a sense of racial difference and inequity that persists today. In the modern United States, for instance, Black women are sometimes three to four times more likely to die from issues during pregnancy compared to white women.[9]

Experimentation was wide-spread among medical students and professionals alike. In the 19th century, many American medical schools used Black bodies stolen through grave robbing for dissection and medical experimentation.[12] This thought process of viewing Black bodies and persons, alive or dead, as an "other", ripe for experimentation, persisted well into the Jim Crow era. For instance, the Tuskegee Experiment, initiated in 1932 and ending in 1972, infected Black men with syphilis against their knowledge, leading to over 100 deaths.[13]

This experimentation went hand in hand with a legacy of colonialism. Colonies served as ample testing grounds for medical advancements by European and North American medical professionals.[14][15] Honor Smith, a British physician at Oxford University, was jovial about the opportunity to test on colonial populations; in 1955, he stated, “[I]t is the almost unlimited field that Africa offers for clinical research that I find so enthralling…problems of the first interest abound, [and] clinical material is unlimited."[16]

These experiments coincided with developments in anthropology and theory around race, with many prevalent thinkers of the time believing there to be distinct differences, even distinct origins between different races.[6] After World War II, scientific racism and eugenics began to be dismissed at large by the scientific community. While certain experiments and explicitly eugenicist thought continued, many of the theories surrounding racial difference and eugenic superiority were discredited and pushed out of the mainstream.[17] Certain medical practices such as equations and decision-making tools continued with these biases in mind. As the internet developed, diagnostic tools were often based on decision-making tools embedded with a perceived racial difference.[18] In the 21st century, these tools and methods have come under more scrutiny.[18]

Over the past few decades as our understanding of race and ethnicity has evolved, advocates for change have been questioning the assumptions that numerous medical decision making tools and formulas are based upon.

In recent years, students, petitions and social media campaigns have recently been at the forefront for driving medical institutions to change the way they utilize certain clinical decision making tools such as eGFR calculations.[19] In 2017, Beth Israel eliminated the race factor in calculating kidney function.[18] In May 2020, through medical student advocacy to their administration, the University of Washington transitioned to a new eGFR calculation that excludes race as a variable.[20]

The estimated glomerular filtration rate (eGFR) is supposed to give doctors a sense of how well the kidneys are working as they separate out waste from the body. Algorithms often differ between racial groups in the United States, with Black patients receiving generally higher eGFR values than their white counterparts.[21] The eGFR remains the standard across the United States, with professionals citing higher serum creatinine concentrations among Black patients compared to white patients.[22][23] Reasons for this assumption vary, but it is thought to partially stem from the notion that Black persons are more muscular.[22] This idea has come under more and more scrutiny, with certain institutions choosing to deviate from the race-adjusted model as a result.[24] These biases have sometimes delayed or complicated medical care, a factor that exacerbates already higher rates of death kidney failure among Black persons compared to the overall population.

Kidney transplants also have racial implications, both direct and indirect. The Kidney Donor Risk Index (KDRI), the United State’s official kidney allocation index, was developed in 2014.[25] Race is among the factors used to predict the success of a kidney graft, with Black donors’ kidneys often thought to perform worse than kidneys from other donors.[26] Being Black results in a demarcation as a less preferable donor by the KDRI.[27] This creates a snowball effect, with less kidneys from Black donors in the system. In turn, Black patients in need of kidney donations are affected. Black patients already face longer wait times than patients of other races in need of kidney transplants. Black patients are more likely to receive a kidney transplant from a Black donor, according to recent studies.[27] This lack of resources can exacerbate the already lengthy wait times. In recent years, some professionals have called attention to these disparities, advocating to replace “the vagaries associated with inclusion of a variable termed ‘race’.”[26]

In the United States, African American and Hispanic women continue to have higher rates of cesarean section than do white women in the US.[28] Medical decision making formulas such as the Vaginal Birth after Cesarean (VBAC) algorithm have been found to contribute to such disparities for women of color. The VBAC is used to estimate success rates for vaginal birth among expectant mothers with histories of previous cesarean delivery and is used by clinicians to counsel whether to offer trials of labor or repeat cesarean section.[29] There are two race-based correction factors incorporated in the formula, one for African American women and one for Hispanic women. These factors are subtracted from the likelihood of having a successful vaginal birth after Cesarean and thus inherently predict a lower risk of vaginal birth success for African American and Hispanic individuals.[30] These lower estimates may discourage health providers from offering trials of vaginal labor to expectant women of color, although the health benefits of vaginal birth (lower rates of surgical complication, faster recovery, fewer complication in later pregnancies) have been well documented.[1] These race/ethnicity associations have been challenged by health providers since they have not been thoroughly supported by biology and are concerning as black women have higher rates of maternal mortality.[31]

Spirometers are devices that measure lung volume and airflow. They are used for the diagnosis and monitoring of several respiratory diseases such as asthma and chronic obstructive pulmonary disease (COPD).[32] In the United States, spirometers use “correction factors” for individuals labelled as Black (10-15%) or Asian (4-6%).[1] There have been several concerns about equity in these lung function formulas including the misdiagnosis of disease severity in asthma and COPD.[33] Recent interest in investigating the basis for why these correction factors came to be spawned a systematic review of 226 articles published between 1922 and 2008 that found that less than one in five studies defined race and that researchers frequently assumed inherent or genetic differences.[34]