Listen to a discussion of this article through our AI-generated podcast.
CAP TODAY now offers the option to listen to a discussion of our articles with voices reminiscent of a Siri or Garmin speaker. In fact, the voices that you will hear are completely AI generated. Since this AI-generated podcast uses technology that is still in its early stages of development, there may be some errors in pronunciation and the tone may sound too upbeat for the material covered.
Karen Titus
October 2024—Using a race adjustment in maternal serum alpha-fetoprotein screening has physicians sitting on the fence these days.
Including race in calculating risk for open neural tube defects has been a longstanding practice in medicine. Adjusting for higher rates of AFP levels seen in Black pregnant patients, proponents say, allows this population to receive equitable care.
First of two parts.
In the November issue, a mix of laboratory views
That premise lies on one side of the fence. On the other are those who maintain the practice is suspect, even harmful, and that the routine use of a race-based adjustment should stop.
With the experience of dropping raced-based adjustments to estimated glomerular filtration rates still fresh in many minds, physicians are now deciding whether to keep or drop the adjustment for maternal serum AFP, even ahead of any potential changes to guidance by groups such as the CAP.
Little wonder. Because as anyone who has ever sat on a fence knows, it can easily become a shaky, even painful, perch.
The maternal serum AFP discussions often lean heavily into the eGFR/renal function turnabout.
The University of Washington was a guiding voice in those discussions nationally, as “one of the pioneering institutions to think about race as a social construct in the context of laboratory adjustments,” says Shani Delaney, MD, professor and fellowship program director, Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, UW Medicine.
It’s unsurprising, then, that UW has continued to look at the role of race-based variables in medicine. Toward the end of 2022, it dropped the race adjustment for maternal serum AFP. Though the laboratory no longer performs this testing in-house, the outside laboratory partner “has been great in understanding our desire as an institution not to incorporate race adjustments,” says Dr. Delaney, who recently spoke with CAP TODAY about how her Ob-gyn colleagues are grappling with the implications of using maternal race in this screening.
It’s hard to separate AFP from a broader context, she says—which is one of the reasons the conversations can feel so disquieting. As society as a whole and the medical field were deeply evaluating the events of 2020, Dr. Delaney says, it led to “an increased desire to understand the effects of race and racism in medicine.” Many of her UW obstetrics colleagues identified AFP screening as the next logical practice to consider, she says, and decided to evaluate whether the race component was the correct adjustment to make. A deep dive into the institution’s own data helped nudge them toward their decision, says Dr. Delaney.
She and colleagues performed a retrospective analysis of patients who underwent prenatal analyte screening between January 2007 and December 2020 (Burns NR, et al. Obstet Gynecol. 2023;141[3]:438–444). Of 43,997 patients who underwent analyte screening, 27,710 patients had complete data for analysis, six percent of whom were identified as Black. New multivariable linear regression models were created to evaluate the relationship between gestational age, maternal weight, and maternal race on AFP levels, which were compared with laboratory-derived calculations that used historically determined race adjustments. The researchers report identifying no difference in values between Black and non-Black pregnant individuals when adjusted by maternal weight and gestational age at blood draw.
The goal wasn’t to drop the adjustment. “We didn’t approach this already knowing what we wanted to do,” Dr. Delaney says. “But we wanted to have an intelligent and thoughtful conversation about it.” That led them to dig into meanings behind the math. She explains: “We wanted to focus on race in prenatal screening, not just statistics.” Much of the recent literature and responses to it, she notes, have been arguments about math: “Are we using the right statistical analysis? Are we adjusting correctly? Is that the right multiples of the median? It’s focused on the statistical interpretation of the data.”
“I think the question is actually a much bigger one,” she says. “It’s really about the understanding and approach of racism in medicine, and using race, which is a social construct, as a proxy for genetics and biology. And that’s not a statistics question.” Rather, she says, researchers need to incorporate societal and scientific perspectives as they evaluate the gaps in research and other types of adjustments that might be needed.
Response among Ob-gyn colleagues at UW has been overwhelmingly positive since the race adjustment was dropped, Dr. Delaney says.
When they first began considering the change, one of the questions they raised was how patients’ race had been determined previously. Though patients were asked to self-identify their race at some point when they were registered into the medical record system, “We know anecdotally that doesn’t always happen,” Dr. Delaney says, adding that sometimes providers, including clinicians or laboratory staff, will assign a patient’s race based on their own assumptions. “This is one of the challenges of using a social construct as a scientific variable—you need to understand how that is defined,” she says. And, as the Burns article notes, older studies looking at maternal serum AFP have used shifting, imprecise terms to define patients, including “Black,” “African-American,” and “Afro-Caribbean.”
With the race adjustment now gone, “There has been a good amount of satisfaction in eliminating that uncertainty of having to ask a patient the awkward and inappropriate question of, Who are you? As if somehow that’s supposed to be related to a biological variable in a lab test,” she says.
She and her colleagues were adamant in their arguments. Inserting race as a scientific variable, Dr. Delaney says, “perpetuates a racist approach to race adjustments. We want to make sure that we are not conflating someone’s personal identification as they socially define themselves and equating that with their blood work.” In addition to putting patients more at ease, “We were appropriately using race as a social construct and not as a genetic variant.”
Race shouldn’t be eliminated as a factor in clinical medicine or medical research, she’s quick to add. “But it’s about using race appropriately.
“It’s absolutely essential,” she continues, “to understand how clinical outcomes or laboratory tests are different or vary by social groups,” including, potentially, race, insurance status, language, and sexual orientation. By understanding differences in social variables, “that’s how we tackle and eliminate health inequities.”
But, she says, race can’t be used to explain differences in clinical outcomes—even if those differences are due to race. “It’s how we use race, and how we understand health inequities, and not how we use race to explain biologic variables.”
“It’s an important distinction,” Dr. Delaney notes, “but it’s often confusing.”
Confusion, as well as conflicting data and research gaps, is keeping many on the fence.
Everyone shares the same goal—to provide the best, safest care possible for patients. Shouldn’t that include adjustments of risk estimates to improve screening accuracy? Dr. Delaney acknowledges the concern, which is: We don’t know what they mean, but there are differences between groups of patients. We should use those adjustments until something better comes along that will explain it better.
She’s familiar with the arguments. The Burns, et al., article in Obstetrics and Gynecology and a clinical opinion (Pierre CC, et al. Am J Obstet Gynecol. 2023;229[5]:522–525) have garnered such responses, which have suggested that race should continue to be used for adjustments because mathematically they “work”—you can hear the air quotes as she speaks—and that the “why” is less important.
The problem in continuing to use a race adjustment—even if everyone agrees the practice is imperfect and problematic and should be on its way out the door, if not now then eventually—is that it continues to perpetuate social injustices and racism, she says. “It’s not acceptable to use race as a surrogate, as sort of the ‘easy’ way out, rather than taking the right way out.”
Dr. Delaney suggests it would be helpful to move beyond this ongoing back-and-forth discussion. “We have an obligation as physicians and scientists to take that bigger step back and not accept that as an okay rationale for why we should continue those race adjustments.”
As she and others note, the literature itself is conflicting. There are indeed maternal serum screening studies that identify differences in racial groups. But, Dr. Delaney says, it’s important to ask about the biological reasons for those differences. Among the possible answers: environmental exposures that are more common in certain racial groups based on where they live, and based on exposures during pregnancy; the effects of racism and stress on placental attachment to the uterus and the relation to maternal/pregnancy hormones; and differences in medication exposures by race.
Again, these are biologic reasons that can vary in different racial categories, she notes. “But it’s not race per se. It’s the exposures that happen due to structural racism and/or different exposures by race.” Once the biology is better understood, then and only then should physicians “appropriately adjust for those in our prenatal screening tests.”
Is there much appetite for such research? Dr. Delaney thinks there is, especially in the genetic screening world. But funding is another matter. “I would hope there would be enough scientific interest that people will begin to explore biologically plausible mechanisms for potential differences in race.” It will take dedication to dig into it from the scientific community, she says. Whether there will be funding for people to perform studies on a large enough population is another matter, she adds, noting that the monetary value of the test also plays a factor in how research is funded.
Dr. Delaney brings personal as well as a professional experience to the topic. She herself is multiracial. “Where would you put me?” she asks with a laugh. The race adjustment for Black versus non-Black “does not work in the reality that we live in in the United States, or the world.” The so-called easy way is, she notes, anything but easy.
The journey at UW was carefully thought out by many groups, Dr. Delaney says, including clinical laboratory directorship, genetic counselors, and maternal-fetal medicine providers.
All three groups were a vital part of reviewing the literature and evaluating their own data before making policy changes. “Everyone felt like they were part of the process.” Throughout the collaboration, they kept a larger picture in mind. This is not a laboratory decision; it’s a medicine decision. Moreover, it’s a decision about medicine in society. Not only did Dr. Delaney and colleagues review the medical literature, she says, but also the concepts of structural racism and how it affects medicine as a whole.
It can be a hard sell for medicine, she acknowledges. She gets it. “It can be anxiety producing to clinicians and scientists to not be able to quantify something,” she says.
But, she adds, that can—and has—veered into shameful territory. “It has shades of the one-drop rule, right?” she observes. “How Black is someone? How Asian is someone? How Hispanic? How do I correct for that? And that’s not a value that we can put on someone.”
That’s why additional research about biologically plausible mechanisms that differ by race is the more useful avenue to pursue. “So we can quantify those in the correct way.”
This should not be a conversation solely about math in medicine. “We can’t put a mathematical value to race. This isn’t about what is quantifiable in that way,” she says. “It’s about how we need to redirect our overarching approach.”
What this actually comes down to is, what is ethically, socially correct? What is right from a justice perspective, for us not to incorrectly use a variable that perpetuates racism?
“It’s hard to step away from that,” she concedes. The history of using race in medicine is a long one, and history doesn’t loosen its grip easily. “And the work that needs to be done is hard. Doing the correct thing is harder than using a longstanding ‘correction.’”
She’s familiar with physicians’ concerns that removing the adjustment would undermine long-held practices, not to mention recommendations from professional societies, and lead to harm.
But as she and her colleagues advocate for removing the adjustment soon, and not at some hazy point on the sooner-or-later continuum, they raise a different concern about patient well-being. “We need to acknowledge it causes harms as well,” Dr. Delaney says.
“We all want to provide excellent care that meets or exceeds the standards of society,” she adds. That’s not up for debate. Instead, she proposes a question to those who find the adjustment useful. “Can you be open to the discussion that our current standards are based in inappropriate use of race? Until we can really tackle that as the primary question,” she says, “I think it’s hard for us to move forward.”
But she sounds confident that change will come. “Change takes time. It takes bravery. It takes insight and introspection. And that’s something that comes a little more slowly than we sometimes hope it would. But I think with persistence and ongoing education, we’re heading in the right direction.”
Karen Titus is CAP TODAY contributing editor and co-managing editor.