A recent study by Michigan Medicine researchers found that pulse oximeters — small blood-oxygen reading devices that clip to the finger — were found to be less accurate in Black patients than in non-Black patients. These findings add to mounting evidence that pulse oximeter technology can be biased with regard to skin tone, a reality with disproportionate racial consequences.
Valeria Valbuena, a general surgery resident at Michigan Medicine, said she was motivated to lead the study after reviewing work done by her U-M colleague Michael Sjoding and others, who found that racial bias exists in pulse oximeter readings in Dec. 2020. The 2020 study was cited by the Food and Drug Administration in an announcement warning patients and health care providers of the inaccuracy.
According to Valbuena, the 2020 article, which was published in the New England Journal of Medicine, received mixed reactions. Valbuena said this paper, like other research on the topic from the 1990s and early 2000s, was limited because the results could not be confidently generalized to larger populations.
“That’s when we started thinking: Were there any opportunities to replicate these findings in a different patient population while avoiding some of the limitations of the previous papers?” Valbuena said.
The recent study analyzed data from 372 patients pulled from the Extracorporeal Life Support Organization, a clinical registry of patients who experience lung failure and subsequent use of extracorporeal membrane oxygenation, a heart-lung machine.
Each of these patients had received both pulse oximeter readings and arterial blood gas tests. According to Valbuena, the arterial blood gas test is more accurate in measuring blood-oxygen levels, but the process is more invasive and takes longer to perform.
Valbuena said the study included patients from diverse ethnic and racial backgrounds, allowing the researchers to make comparisons. After analyzing the data, Valbuena said they found a significant discrepancy between Black and non-Black patients.
“If I have 100 patients, how many of them are going to have hypoxemia with the pulse ox reading normal and the blood-gas not being normal? And that was about 10% for white patients, and twice as much for Black patients,” Valbuena said.
Pulse oximeters are frequently used in caring for patients hospitalized with COVID-19. The pandemic has disproportionately affected people of color, especially Black communities. Racial bias in pulse oximeters became more visible as a result of the pandemic’s disparate effects on people of color.
Victor Agbafe, an MD/JD candidate at the U-M Medical School and Yale Law School, said the results of the study should encourage scientists to understand why these essential medical tools are exhibiting a racial discrepancy.
“I was alarmed more than anything else,” Agbafe said. “Given that pulse oximetry is a key tool that we use in medical triaging, I felt that it was extremely important that we have a sense of urgency to see why these devices are less likely to catch hypoxic conditions in Black patients.”
Valbuena said pulse oximeters measure blood-oxygen levels via the interaction between two lights on the device and the hemoglobin in blood. Due to the light detection process, Valbuena said higher rates of melanin in the skin may play a role in how the pulse oximeter reads blood-oxygen levels.
“There are a number of things that affect how the light interacts with hemoglobin in the blood, and one of those things is melanin,” Valbuena said.
Bunching people along racial lines, Valbuena recognized, is an imperfect proxy for skin color since skin tones vary widely across people of the same race. Valbuena said Hispanic and Asian patients, who were also included in the study, often have results comparable to white patients.
“In our case, we’re using (race) as a proxy just for skin color, knowing that in a larger sample population, patients who identify as Black or of African American descent tend to have darker skin colors just because of their ethnicity,” Valbuena said.
Valbuena also emphasized that the issue with pulse oximeters and melanin has been known since the 1990s, and the lack of action to solve the issue has disproportionately impacted racial minorities for years.
“I think that communal negligence is a perfect example of how medicine continues to systematically neglect and provide worse care for minority patients, especially Black patients,” Valbuena said. “The convenience of not changing it overpowered the fact that we have had millions of people affected by this phenomenon over the years.”
While the science is strictly tied to skin tone, Valbuena said the issue of neglect is a societal one.
“I feel like if the machine had not been working for white people, things might have been a little different,” Valbuena said.
Agbafe echoed Valbuena, saying he heard from members of the Black community about their lack of trust in the medical system.
“This trend shows why many in the Black community do not trust that our medical institutions have their best interests in mind and this sentiment is one that Black patients in the hallways and elevators have shared with me during my clinical clerkships thus far,” Agbafe said.
Elliott Brannon, Rackham and Medical School student who studies health care infrastructure, said the disparity is also an issue that must be acknowledged at the state level.
“We do need to work to point out specific disparities such as oxygen monitoring,” Brannon said. “But that also needs to connect addressing systemic issues at the level of the state.”
Brannon encouraged students to organize and put pressure on those in power to address the situation.
“Do we need 10 research articles? (To) talk to members of Congress?” Brannon said. “How many people do we need to organize to address this issue and does anyone have enough passion and enough time about this pretty specific topic to make sure it is addressed?”
Valbuena is currently conducting further related research and hopes to attain more findings in the coming months. Valbuena said she thinks the medical community, especially in light of the pandemic, is finally building significant momentum around the problem of bias in pulse oximeters.
In the meantime, Valbuena said the education of medical workers is paramount.
“There’s one of these machines in every single patient room in every single hospital ever. We need to be more aware of the issue so that if someone is struggling to breathe or is showing other signs of not getting enough oxygen, we don’t rely on the pulse ox,” Valbuena said.
A decreased reliance on pulse oximetry, as well as applied pressure on the government and device production industry, are the solutions on the table, Valbuena said.
Where these efforts will lead, Agbafe said, is still unknown.
“I am hopeful for the future as I witness more honest conversations taking place around these issues, and honest acknowledgment is the first step to solving these issues,” Agbafe said. “However, we must judge ourselves on the results of tangible action we take, and the jury is still out on that.”
Daily News Contributor Dominic J. Manzo can be reached at dmanzo@umich.edu.