Understanding, challenging racism in pulmonary function testing

Key takeaways:

• Clinicians need to remember that a patient’s reported race includes social factors in addition to genetics.
• In education, research and reference equations, there are ways to avoid cultural racism.

WASHINGTON — By using race specific equations when interpreting pulmonary function tests, racism and bias come to the surface, according to a presentation at the American Thoracic Society International Conference.

According to Aaron D. Baugh, MD, assistant professor of medicine at the University of California, San Francisco, in an attempt to simplify spirometry data early on, researchers began using an individual’s reported race without considering the social factors that also make up race.

“As we tried to simplify and make things convenient and usable for physicians so that you don’t have this mass of complex data, there was a simplification that said, ‘Well, we can use self-identified race and we’ll capture certainly their ancestry, their genetics, their body proportions,’ with ignoring that you’re also going to be capturing all these other things in terms of the racism that person has faced [and] the unequal social experience,” Baugh said during the presentation. “In this move to simplify and make things more usable, we get this phenomenon of this identification, the racialization of lung function differences for many causes to ‘Oh, it's because of your race.’”

With this type of thinking, bias enters into lung function interpretation for nonwhite individuals and can lead to clinicians misattributing lung function causes, Baugh continued.

“What this can do is that you then begin to normalize the idea that a nonwhite patient should have a lower lung function, or it’s not a problem that, ‘oh, you have this low value of your FEV₁ 70% because that's what I expect for you,’ even if it was really caused by malnutrition or being in a neighborhood with lots of pollution,” Baugh said.

Using an example from a study on COPD in heavy smokers that he conducted with colleagues, Baugh explained how underestimation of disease severity can emerge in patients who are nonwhite when race specific equations are used to estimate lung function.

According to Baugh, centering the margin is one way to avoid misestimating causes of lung function in diverse populations.

“Because we know that with our biases and when we use systems, we can tend to overlook the experiences of the marginalized, making sure you have a healthy sample [is important],” he said. “This was over 500 African American participants, one of the larger studies that allowed me to see that when we look at an outcome and check our assumptions, when we allow those marginalized people to have a voice and when we attend to patient-centered outcomes, we can now see that maybe the way we were doing lung function before didn’t work.”

In a study by Gabrielle Y. Liu, MD, and colleagues that evaluated racial differences in emphysema, Baugh further highlighted the harm race specific FEV₁ can have on nonwhite patients. After adjusting for several variables, including age, current smoking status and pack-years history, Liu and colleagues observed that emphysema incidence was around three times unexplained in Black patients compared with white patients, Baugh said.

He added that this difference between white and Black patients was smaller when researchers used a race neutral FEV₁ equation.

With these study results in mind, it is important to consider the instances where lung function is used, Baugh said. This includes disease classification, medical eligibility, surgical safety/eligibility, disability evaluation, employment/pre-employment screening and public health measures, according to Baugh’s presentation slides.

Baugh concluded his presentation by explaining ways to avoid cultural racism in education, research and reference equations.

“In education, making sure that we message to providers that they need to think about the broader patient context,” Baugh said. “It’s not just one number that gives a solution; even though that simplicity and being able to economize is appealing ... we still have to be clinicians and look at the whole context of a patient.”

In research, Baugh said more diverse samples and consideration of different, relevant exposures are needed.

“Ultimately, we want to use reference equations that don’t give in or lend to this phenomenon, ordinariness or invisibility,” Baugh said. “We don’t want to be having biases that hide things from us. We want to see things in formats that allow us to visualize, that force us to think hard about what really is the cause of the lung function decrement we’re seeing in this patient and whether it correlates with their symptoms and their clinical course or not.”

References:

• Baugh AD, et al. Am J Respir Crit Care Med. 2022;doi:10.1164/rccm.202105-1246OC.
• Liu GY, et al. Ann Intern Med. 2022;doi:10.7326/M22-0205.