Is Patient Race Important -- Or Just Convenient?

F. Perry Wilson, MD, MSCE


March 18, 2020

This transcript has been edited for clarity.

Welcome to Impact Factor, your weekly dose of commentary on a new medical study. I'm Dr F. Perry Wilson.

I'm coming at you from my home, as the Yale Broadcast Center is closed because of the coronavirus outbreak. Apologies in advance for the video quality, and for the screams of young children who have — let's be honest — just no respect for in-depth discussion of medical studies.

This week, a bit of a diversion from the latest virus news to talk about the other elephant in the room, one that has been around for a lot longer.

I'd like you to think about the last time you gave a formal patient presentation. You know, for a talk, during rounds, or whatever. You probably said something like: "This is a 38-year-old black man who presented with a chief complaint of fever," and it went on from there. Age, race, gender. I know that style was engrained in my head. But... why race?

I got to thinking about this when I read this research letter, appearing in JAMA Internal Medicine, talking about our eGFR estimating equation.

GFR, or glomerular filtration rate, is an important metric for kidney function, but it's really hard to measure directly. It requires a timed infusion of a drug like iothalamate and pretty advanced analytics; you're not getting this at your local lab.

So, brilliant scientists (some of whom are my friends and colleagues) developed estimating equations that can take just a few easily measured variables and predict eGFR. The best such equation, known as CKD-Epi, looks like this:


It requires knowing four things: the patient's creatinine level, age, sex, and race.

There's race again. Why does race matter? Why is it so central to our understanding of our patients?

In the context of GFR estimation, I was taught that creatinine comes from muscle, so people with more muscle have a higher creatinine value at the same GFR. And, on average, black people have more muscle, so we need to account for that.

We are using race to infer something biological. What I never thought to ask is, well, if we are accounting for muscle mass, why don't we just measure muscle mass?

But the truth is, race doesn't really account for much biologically. Race is a label we assign on the basis of skin color. That is a biological factor, but so is eye color, hair color, toenail length. Why has race been given so much credence? Why is it the second thing we say about a patient?

If your answer is that it is a proxy for a variety of genetic differences that do have medical relevance, well, it doesn't seem to work that well. Allelic variation is greater within a race, as typically defined, than between races.

What about kidney disease? Well, that eGFR estimation is really important to nephrologists like me. You can't get evaluated for transplant, for instance, until the eGFR is below 20 mL/min/1.73 m2. We might think about starting dialysis when the eGFR is below 10 mL/min/1.73 m2.

Given the same age, sex, and creatinine value, someone who identifies as black will have a higher eGFR than someone who identifies as white. Take this hypothetical individual.


If he describes himself as white, he is eligible for a transplant. If he describes himself as black, he isn't.

But what do the data show? The article compares the measured "true" GFR of 2601 self-described black individuals with the estimated GFR, with and without the use of race in the estimating equation. You can see the results here.


Without using the race coefficient, the eGFR is systematically too low. Using the coefficient, the estimate is better, especially in the range south of 75 mL/min/1.73 m2, which is what we tend to focus on.

Using height and weight instead of race did not rescue the accuracy of the equation.

So does this show that we need race?

Not really. It just shows that race, which is a weak reflection of some biological stuff relevant to GFR, is better than height and weight, which are weaker reflections of some biological stuff relevant to GFR.

Why shouldn't we use race? The main reason is that the use of race in medical models or for medical therapy amplifies the idea that there are relevant biological differences between the races, that race may dictate response to a variety of interventions.

In other words, we see that race is in a model and conclude that race is important as opposed to the reality, which is that race is convenient.

Of course, black individuals do have worse outcomes in various disease states. But with rare exceptions, it's not because of genetics; it's because they are exposed to the systemically toxic effects of racism.

So instead of putting the convenient race term in our models, we need to invest in the measurement of those social factors that really drive the differences in outcomes: poverty, isolation, poor education. Then we can know what to try to fix.

In other words, maybe our case presentations should begin: "This is a 38-year-old uninsured man with a chief complaint of fever."

F. Perry Wilson, MD, MSCE, is an associate professor of medicine and director of Yale's Program of Applied Translational Research. His science communication work can be found in the Huffington Post, on NPR, and here on Medscape. He tweets @methodsmanmd and hosts a repository of his communication work at

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.