Note: yes, this post will eventually circle around to science, I promise!
A very close and beloved family member recently asked me, “Do you think that young people just haven’t had anything else to protest in their lifetime?” When I echoed this to my husband later, he cringed, a sign that yes, this is an objectively ignorant thing to say. When I responded to this family member, who I love so very much, I took the opportunity to do what I thought was building a bridge instead of starting a fight. In doing so, I regret not speaking my mind. I told this family member that I thought, yes, I had heard/read that this enraged response to the death of George Floyd and many others might be amplified because of the pandemic keeping everyone at home, stir crazy, and starved of action. Not what I wanted to say.
What I wish I could have said to her, had I been able to find the grace and patience to respond without inflammation, was no, I don’t think that is true. I’ve witnessed lots of marches and protests in my lifetime for all different types of causes. So that tells me that not everyone is paying attention. This particularly horrible issue of racial inequality is finally getting the attention it needs. I’m going to venture to say that I don’t think this is a political issue; it’s a fact. People of color are disproportionately affected by police brutality, and that is just not something that I or lots of other people can live with.
I am not brave enough to go out and stand at the protests with my brothers and sisters, but I am grateful you are doing this because it is making a difference. The way I believe I can make a difference is by putting my words out there and hoping that someone will find them meaningful. So here is a post that I actually wrote a very long time ago and didn’t have the guts to publish because I thought it might label me and subsequently push away some of my potential clients, family, and friends. I hope that those who read this can keep an open mind if this perspective doesn’t agree with yours.
I wrote this post in response to a book I read that really changed the way I think about racial issues. Before reading the book, I had seen a Facebook post from my aunt who had attended a talk called “White fragility: confronting underlying racism.” In her post, she mentioned a book with a similar name. I was intrigued, so I checked the book out from the library. No matter what race you identify with, I highly recommend reading it if you are at all interested in becoming a better person!
Briefly, the book talks about how racism is still very much a part of our society (duh), but not in the ways you may think. It is buried deeply in our institutions and the way we do things in America, and claiming that we are not racist as a group (or even as individuals) is exactly what perpetuates this problem. Because we are human, every one of us is susceptible to implicit bias. The good news is that we can work against this bias by first becoming aware of it.
I began to think about how racism might be built, however unintentionally, into science and medicine. There are some obvious ways. For instance, most experts acknowledge that there are very real racial disparities in healthcare. In other words, access to quality healthcare is not available equally to people of all races, whatever the reason. There is also a historical context for distrust of healthcare providers, possibly stemming from horrible instances in history such as the Tuskegee Study. But after thinking for a while, I realized that there are some less obvious ways that races are not treated equally in science.
In the clinical trials I conducted during grad school, we measured and reported demographic characteristics, including race, from each study. Aside from assessing differences in the study results by race (which we never did), I learned that reporting the racial makeup of the sample is done in part to make the case that the study sample represents the general population as closely as possible (and therefore, we can generalize the results to the rest of the population). Almost invariably, between 70 and 80% of our participants were white, while maybe 10% were black at most (more often it was closer to 7-8%), and the remaining 3-10% were a smattering of “Asian/Pacific Islander” lumped into a group, and everyone else falling into “other.” I thought about these numbers long after I had published my results and wondered if the racial makeup did deem the study generalizable. When I searched, I found a recent study that looked at this very thing and found that, in the clinical trials for FDA-approved drugs between 2013 and 2015, only 6.4% of participants were black or African American. On a national scale, that’s not good enough to represent the census-measured 13.4% of the US population that is black, not to mention the other racial designations. And we haven’t even gotten into ethnicity yet.
Maybe more importantly than the proportions matching those of the US population, in my opinion, is that researchers actually look to see if the results differ by race. Some do this, but most don’t. To do this, we need to have adequate statistical power. That means that if there aren’t enough people of Asian descent enrolled in the study to compare them statistically to white participants, then we won’t find out if there’s a difference. Yes, this might mean that the study will cost more if more participants are needed overall, but the advantage is that we are going to get closer to the truth in our research and that people of all colors will be better treated when they have a medical issue. For those of us that are more money-minded, consider the advantage of learning which demographic groups benefit the most from a drug, paving the way for more effective marketing.
I have another issue with how we predict disease based on risk factors. Race is very often included as a risk factor for chronic diseases. The first time I realized I had a problem with this was during a lecture in grad school. It was actually an undergraduate nutrition class for which I was a teaching assistant. One of the PowerPoint slides discussed the various risk factors for a disease (maybe it was type 2 diabetes or hypertension), and it mentioned that black people were more likely to be affected by this disease. I thought for a second, and I realized that this seemed kind of ridiculous. Sure, it might be true, but that makes it sound like black people get the disease more often because they’re black. I know that not to be true because of my knowledge of what “increased risk” means and that it does not mean “cause” in any way, but it still seemed fallacious to me. When I brought this up to my advisor who taught the class, she said it could be that socioeconomic status is conflated with race and that race is designated as the risk factor because it’s easier to measure, whether in research on a demographic questionnaire or in a doctor’s office by visually profiling a patient. If that’s true, then socioeconomic status might be a better predictor of developing some types of chronic disease.
If this is confusing, think about it this way: do black people of a high socioeconomic status still get the disease more often than white people of the same high socioeconomic status? If not, then race is not the best predictor of the disease.
Here’s the problem I have: even though we in the scientific community know that risk factors such as race are not necessarily causes of diseases, sometimes we treat them as if they are. Patients of color may feel that they are destined to develop a disease, and this may affect their habits and outlook on life. This idea is explored extensively in some of Malcolm Gladwell’s works, including Outliers (which I highly recommend!).
The solutions I propose:
Let’s call it like it is and get better at identifying the right risk factors. If socioeconomic status is a better predictor of disease than race, then let’s get better at measuring socioeconomic status. If there are other cultural factors that are often tied to a race but better explain what we see, let’s figure out what those factors are.
Let’s also actually enroll participants that racially and ethnically represent the population and then look for differences by race in each of our studies, the same way we analyze results by sex or gender. More importantly, we can think a little more deeply about the data we are collecting and what it really means; how valid is our measure of race, really? If someone chooses to answer a race question as “other” because they are half white and half black, how good are the data? Every time we scientists publish a paper, we have a choice in how we present our results. We can choose to do it the way it’s always been done because it’s easier to get it through peer review, or we can be true pioneers in making science better and try something new.
P.S. I have also been a participant in research studies and numerous surveys of all kinds (I’m sure all of us have been at some point). Why are there only three or four race categories? People who identify as white do not all come from the same cultural background, and the same goes for people who identify as black or Asian. If we’re going to spend time classifying people like this, why don’t we try to get a little more accurate?