SANTA CRUZ – For years a growing chorus of research has suggested that pulse oximeters, the clip-on blood oxygen measurement tool made ubiquitous by the COVID-19 pandemic, can produce inaccurate readings in darker-skinned patients.
But experts say that until recently, little was known about the specific adverse clinical outcomes this potential bias could produce.
Analyzing data from the COVID-19 pandemic, a study from Sutter Health found that the pulse oximeter’s inability to precisely read blood oxygen levels in Black patients could have led to 4.5-hour treatment delays, according to a release from the provider.
Oakland-based Roots Community Health Center and UC San Francisco were collaborators in the research effort.
Study representatives recently told the Sentinel that because COVID-19 treatment guidelines put the non-intrusive fingertip device at the forefront of detecting the threshold for clinical action, its limitations may have had real-world consequences.
One of the lead authors of the paper is Sylvia Sudat, an accomplished research scientist and biostatistician with Sutter.
Sudat was raised in Santa Cruz County, graduated from Harbor High School and spent much of her childhood in and out of the lab in Scotts Valley, where her mother was a clinical laboratory scientist.
“I felt closely connected to health care, but also saw how hard it could be to navigate, even with a family member working in health care,” Sudat told the Sentinel, later adding that “the best care is equitable care.”
Sudat, a senior biostatistician and scientist at the Center for Health Systems Research as well as a core member of the Sutter Health Institute for Advancing Health Equity and the Data Science team, said that while the device systematically overestimated blood oxygenation by 1% more in non-Hispanic Black patients, that seemingly minor difference carries profound implications.
“I’ve had people say to me ‘Well, that’s not a big deal, why are you getting all excited about that?’ ” Sudat said. “When we’re talking about health care and people’s survival, small systematic differences do matter and we need to take them seriously.”
In addition to treatment delays, the overestimation in Black patients, according to the Sutter release, was also associated with:
• Increased time and lowered probability of receiving dexamethasone, a drug used to treat COVID-19.
• Lower hospital admission probability.
• Lower probability of receiving supplemental oxygen treatment.
Sudat explained that the device detects oxygen levels by shining a light through the finger. However, Melanin – a natural pigmentation – absorbs some of that light, which systematically shifts the reading. The more Melanin the darker the skin tone.
“Because it’s kind of a one-size-fits-all device, it doesn’t mean the same thing,” Sudat said. “It’s not just that it’s more variable or there’s just a greater amount of inaccuracy, it’s (the measurement) actually moved upward and so basically the translation between what you see on the pulse oximeter and what’s actually going on in your blood is different.”
Kristen Azar, a senior author of the study, said Black patients also experienced a higher frequency of arterial blood gas tests, which is the gold standard for measuring blood oxygen levels but is invasive and oftentimes painful.
She said clinicians might order the test as a confirmatory tool, especially if symptoms the patient was exhibiting differed strongly from the pulse oximeter reading.
“It was harder for clinicians to appropriately treat those patients when they were suffering from COVID-19 and it resulted in measurable treatment delays,” Azar said, “and so that contributes to structural racism in health care, regardless of the clinician’s intent.”
The Sutter study focused on patients outside of Santa Cruz and only those presenting COVID-19 symptoms in the emergency department specifically.
Still, the device is commonly used in a number of medical settings and Azar said the study has important implications when it comes to unconscious bias in health care technologies and the industry as a whole.
“Awareness, awareness, awareness,” Azar said, when asked about the practical uses of the study. “If you are going in and this device might not mean the same thing for you as for others, then just make sure you advocate for yourself. Same thing for clinicians.”
Evidently, awareness is spreading.
Last month the Food and Drug Administration convened a panel of experts to discuss “ongoing concerns” related to the device.
According to a report from the Associated Press, industry experts are hoping the meeting will result in updated testing guidance from the agency with study criteria and specific requirements for inclusion of individuals with diverse skin pigmentation.
The AP also reported that some companies selling pulse oximeters have disagreed with the notion that the devices are less accurate for patients of color. One CEO from the medical device company Masimo said an internal review found no evidence of significant differences in accuracy for Black patients.
Still, Azar and Sudat said evidence is mounting and this could just be “the tip of the iceberg.”
“We can’t sweep these differences that we find under the rug, just because we’ve decided that they’re unimportant,” Sudat said. “We need to actually take them seriously and make sure the devices we have are enabling our clinicians to care properly for patients regardless of what their skin color is.”
