Following are remarks I delivered at the North Carolina Nurses Association Annual Conference on September 23, 2021:
Thanks to Laura for that perspective, and for setting the stage for my presentation which will be different in both focus and tone. Why different? First of all I was born in 1951, into American apartheid and was forced to receive inferior healthcare services—when they were available to people of my color at all—through segregated and intentionally underfunded institutions until I was a middle teen. Over the past 50 or so years I have observed an amazing transformation of American healthcare and medical practice in general except for one notable feature: the persistence of racism and extreme disparities in health outcomes for non-white people. I was myself a recent victim of this pernicious and seemingly ineradicable feature of American healthcare back in February 2019. I won’t bore you with the details of that experience, but it is recounted in an essay published on my blog, HBs Fearless Trauma Chasing. You can read it at hbsfearless.com. Just scroll down to the post, “My Story.” You may also want to read the essay “Doctors and Relationships.” That too is a good one.
But back to the matter at hand…
For those of us who experienced legal segregation and overt racial oppression, it was clear that Black suffering was not just a matter of individual prejudice. The social status, control, and abject condition of millions of Black people could not have been maintained by random individual behavior and attitudes. No social system can. It required the active support of the state, of norms that were enforced by legislation, by judicial action, by police power, by economic and political power, by everyday practices of business and industry. Healthcare is a business, an industry. As Laura has shown, it has a very clear and deeply embedded history of prejudice against and contempt toward the Black body that correlates with the dubious racial history of the country in general. So in attempting to locate the source of racial differentiation in healthcare outcomes, it is actually distracting talk about individual prejudices. You have to look at the system and its heritage; you have to look at fundamental practices and assumptions; you have to look at the tools, the processes, the business model.
I want to begin my talk with this simple postulate:
The clinical causes of racial disparity in healthcare are rooted in assumptions, tools, processes and standard practices that people like you employ every day, irrespective of your individual attitudes about race.
You are all practitioners. But none of you created the received information and practices that define your profession. None of you individually designed and prescribed the clinical approaches and operating procedures by which you are bound as a requirement of your licensing. None of you individually created the professional standards and culture by which you are encased as you pursue your craft. Racial disparity in healthcare flows precisely from aspects of those standards and that culture. In other words those standards and that culture are the essence of the industry, the business model—which is why disparity in healthcare—as in so many other of our nation’s industries—is so hard to eradicate.
You say, give me an example of racism in clinical practice.
Well as Laura has revealed, the concept of race itself is a social construct, not a scientific concept. In fact the notion of race—of white and black people as different human types—was invented in this country explicitly to justify Black enslavement. Race is not the equivalent of genetic difference. Yet it is common practice in medical research and clinical practice to “race norm” data and other basic evaluative criteria. A well-known recent example of this bogus, scientistic practice is the debunking of the National Football League’s formula for awarding benefits to players who develop dementia as a result of injuries received playing the game. The clinical “experts” hired by the NFL developed a formula that categorically assumed lower baseline cognitive functions for Black players than whites, meaning Black players would need dramatically lower test scores than whites to qualify for benefits. The result was that Black players were approved for financial awards at far lower rates than whites. Once the bogus methodology and pecuniary rationale for this practice was revealed, the NFL was embarrassed into disowning it.
Less well known is the practice of “race norming” research data, especially in the creation of artificial intelligence tools, such as diagnostic algorithms, and making determinations about the allocation of life saving procedures such as awarding kidney transplants. In each case data derived from non-representative groups of research subjects is arbitrarily “adjusted” by ascribing a “value” to the underrepresented parties, not captured in the original data. Because there is nothing scientific about race, this practice actually represents a corruption of bona fide data, and defeats the claim that this process is “scientific” at all. Although most patients are unaware, their race is routinely incorporated into decision making and formulas doctors use to make diagnostic and treatment decisions. The practice is completely interwoven into American medicine. The effect is usually to disadvantage Black patients. Back in the old days at the dawn of computers and program coding there was a saying, “garbage in, garbage out,” meaning if your programming was based on flawed data, the tool you produced would be equally flawed. Many of the AI tools in healthcare may in fact be derived from deeply flawed data, a fact that is attracting more attention but that the industry may not want to admit and deal with. Meanwhile the “race normed” groups affected by these tools may be left to suffer the consequences of diagnostic error or exclusion from consideration of that needed transplant.
The whole notion of race in supposedly scientific endeavors is totally confounding. The absence of people of color as research subjects in data gathering studies can lead to arbitrary, unfounded adjustments associated with race norming. On the other hand the color of a person’s skin CAN be important in diagnosis and treatment. For example some of you may be aware of the controversy surrounding pulse oxymeters. Those devices detect the oxygen content of blood by measuring light reflected off the skin of a patient. Obviously less light is reflected off dark skin than light skin. The failure to account for this simple fact of physics has led to many Black patients being denied oxygen when their levels were low because white skin served as the basis for the standard calibration of the device. As you can imagine, this flaw had a significant impact during the Covid-19 pandemic, which disproportionately impacted people with darker skins. But this flaw in the tool had been known for years before Covid. The industry simply chose not to do anything about it.
Skin color matters in another way. Certain symptoms present differently on darker skin than on light skin. A rash on white skin may not look the same on darker skin, possibly leading to misdiagnosis. The problem was so bad that a Malone Mukwende a 20-year old medical student from Zimbabwe studying in London, took it upon himself to publish a clinical handbook pointing out the problem, educating the industry to the obvious—that using pale skin as a default in a world increasingly brown and black is not good practice.
And then there are standard clinical assumptions and practices that in many ways just defy understanding. There is certainly no “scientific” basis to them. One example is Body Mass Index, or BMI. The historical origins of BMI are quite clear: It was created 200 years ago during a period of widespread racist theories, not by a physician but by a mathematician seeking to calibrate the perfect human body, and using the European male as his model. Today it is widely believed that the routine use of BMI can be problematic for non-white people.
For some disparity stats there is no plausible explanation except the persistence of a historic attitude within healthcare that devalues and abuses the black body. I refer to the astounding difference in the rate of amputation among white and Black diabetes patients. Black diabetes patients are amputated at a rate THREE TIMES that of white patients. Especially in some areas of the South, the old Confederacy, amputations are said to be so prevalent in Black communities that the population resembles limbless victims of war.
And so now we approach an era of artificial intelligence in healthcare, of algorithms designed from data reflective of a deeply inequitable and flawed system of practice and administration. We edge toward a system both more abstruse and more opaque, not just for patients, but for practitioners as well—people like yourselves. A semi-autonomous system embedded with all the race-denominated flaws that characterize our healthcare system today.
What kinds of reforms might start us on a path of correction and redemption?
Well first, we can compile accurate data on the current extent of disparities in the healthcare system. You will recall that during the pandemic, the collection of such data was random and often non-compliant.
Second, healthcare is a business. Although according to the Commonwealth Fund, which measures such things, US healthcare is the most expensive in the developed world. But the US also ranks dead last in 4 of 5 metrics that determine healthcare quality among developed nations. So we have high profit healthcare with gross disparities. If you are making a lot of money generating disparities what is the incentive to change? You have to change the economic incentives, change the business model. Incentivize equity. To do that there must be among other things, more Black and Brown ownership in healthcare. There is no equity without ownership.
And finally, here is my favorite fantasy reform. Black men have the worst healthcare outcomes of almost any group. Use the health of black men as a proxy for the total system’s performance. The more Black men’s health as a whole improves, the more equitably the system can be assumed to be functioning. Just don’t hold your breath for that to happen… Thanks.
Comments