Race in healthcare is a social category, not a biological one, yet it has been embedded in medical tools, diagnostic calculations, and treatment decisions for decades. The Human Genome Project confirmed that humans are 99.9% identical at the DNA level and that there is more genetic variation within any racial group than between them. Despite this, race has been used as a shortcut in clinical algorithms to estimate everything from kidney function to heart disease risk, sometimes with real consequences for patients of color.
Race Is a Social Category, Not a Genetic One
Social scientists have long understood race as a system of categorizing people based on a small set of physical traits and cultural differences. It was not designed to capture biological reality. Genetic variation in humans follows geographic gradients called clines, meaning allele frequencies shift gradually across regions rather than clustering into distinct racial groups. There are no sharp genetic boundaries between races, no distinct evolutionary lineages, and no specific gene or allele frequency that defines any racial group. It is impossible, for example, to specify a genetic signature that would typify “African ancestry” for 1.3 billion people on the African continent or 45 million African Americans.
This distinction matters because when a doctor’s calculator asks for a patient’s race, it is asking for a self-reported social identity and treating it as though it carries biological information. Self-identification works for cultural and social purposes, but it cannot serve as a proxy for someone’s DNA. A biracial person checking one box instead of another can get a different test result, a different risk score, or a different treatment recommendation.
How Race Got Built Into Medical Tools
For decades, clinical algorithms have used race as a variable in calculations that guide real decisions. Three of the most significant examples illustrate the pattern.
Kidney function: The standard equation for estimating how well the kidneys filter toxins included a “race coefficient” that automatically assigned higher kidney function scores to Black patients. The rationale was that Black adults had consistently higher levels of a waste product called creatinine. In practice, the adjustment made Black patients appear healthier than they were. One study found that removing this race correction would reclassify one-third of Black patients to a more advanced stage of chronic kidney disease. An additional 3% would qualify to begin accumulating priority time on the kidney transplant waitlist. For biracial patients, being classified as Black or white in the algorithm could determine whether their score qualified them for the transplant list at all.
Lung function: Spirometry, the most common pulmonary function test, has used race-specific reference values that require Black patients’ results to be as much as 15% lower than white patients of the same sex, height, and age before being flagged as abnormal. Data from the National Institute of Occupational Safety and Health showed that 94% of Black workers would qualify for disability compensation using race-neutral spirometry standards, but only 81% qualified under the race-adjusted version.
Childbirth decisions: A widely used calculator for predicting whether a woman could successfully deliver vaginally after a prior cesarean section included a race adjustment that lowered predicted success rates for Black and Hispanic women. A study of 302 women found no actual difference in vaginal birth success rates across racial groups. But when race was included in the calculator, about 44% of Black and Hispanic women who went on to deliver successfully would have been told their odds were unfavorable. When race was removed, that number dropped to roughly 10-12%.
Why Race Became a Proxy for Something Else
The reason race appears to predict health outcomes is not biology. It is that race correlates with living conditions shaped by structural inequality. Differences in cardiovascular disease rates across racial groups, for instance, can be attributed to differences in social determinants of health: neighborhood quality, income, access to healthy food, environmental exposures, chronic stress from discrimination, and unequal access to healthcare itself. These factors are rooted in systemic racism and reinforced by ongoing implicit bias in clinical settings.
When a risk calculator uses race as a variable, it is really capturing the downstream effects of these social conditions. The problem is that baking race into the formula treats inequality as an inherent trait of the patient rather than a feature of the system around them. It also obscures what is actually driving risk. Two patients of the same race can have vastly different social circumstances, and two patients of different races can share nearly identical risk profiles.
Clinical Tools Are Starting to Change
Several major medical organizations have moved to remove race from their standard calculations. In 2021, the National Kidney Foundation and the American Society of Nephrology recommended adopting a new race-free equation for estimating kidney function. The updated formula uses the same blood markers but drops the race coefficient entirely, with expanded use of an alternative blood test called cystatin C for more precise results.
The American Thoracic Society, endorsed by the European Respiratory Society, issued an official statement recommending that race and ethnicity no longer be used in interpreting spirometry results. The statement cited evidence that a race-neutral approach is superior for assessing lung health prognosis and the effects of tobacco smoke exposure. Their new recommendation replaces race-specific reference equations with a single race-composite average.
Heart disease risk prediction is shifting too. The most widely used cardiovascular risk calculator, the pooled cohort equations, uses four separate formulas divided by sex and race (white or Black), meaning two people with identical blood pressure, cholesterol, and other clinical values can receive different 10-year risk scores based solely on their racial category. People who identify as neither white nor Black are told to use the white equation. The American Heart Association’s newer risk prediction model moves toward a race-neutral approach, and race-agnostic alternatives like the Framingham Risk Score have been available for years.
Genetic Testing Versus Racial Assumptions
The clearest alternative to using race in treatment decisions is testing patients directly. Pharmacogenomic testing, which analyzes specific genes that affect how a person metabolizes medication, can tell a doctor exactly which drug variants a patient carries. The blood thinner warfarin is a prime example: dosing depends heavily on variants in specific genes that process the drug, and these variants exist across all racial groups at different frequencies. Testing for the actual gene variant is far more accurate than guessing based on a patient’s self-reported race.
Several FDA-approved drug labels already include group-specific prescribing recommendations based on genetic markers rather than race, including medications for seizures, gout, cholesterol, and organ transplant rejection. As pharmacogenomic testing becomes more accessible, race and ethnicity will become irrelevant to these treatment decisions because doctors will have the precise genetic information they need.
What Is Changing in Medical Training
Medical schools are beginning to reform how they teach about race. Traditionally, curricula focused on treating disease with little attention to how social factors shape health outcomes. The consequences of that gap have been measurable: a 2016 study found that medical students and residents held false beliefs about biological differences between Black and white patients, including that Black people have thicker skin and feel less pain. These beliefs directly affect pain management and treatment quality.
Some institutions have developed new coursework that traces the historical origins of race-based medical assumptions and connects them to present-day health disparities. These interventions aim to help future physicians distinguish between race as a social experience that affects health and race as an assumed biological category. The goal is not to stop collecting racial data entirely, since tracking health outcomes by race is essential for identifying and addressing inequities. It is to stop treating race as though it explains biology when it actually reflects the unequal conditions people live in.