In response to complaints that pulse oximeters that measure blood oxygen levels are less accurate in people with darker skin tones, the Food and Drug Administration released Monday long-awaited guidance suggesting how manufacturers should test and label these devices.
The document recommends testing the devices on many more people and provides clearer guidance on the composition of the populations – and of the skin tones – on which they should be tested. Previous guidelines called for 200 data points; the new guidelines suggest manufacturers submit 3,000; they also increase the suggested sample size for pulse oximeter studies from 10 subjects to 150 or more.
Rather than relying on those testing the devices to assess skin tones, the draft guidance suggests manufacturers should use both the newly developed Monk Skin Tone scale and another scale called the Individual Typology Angle (ITA), which is based on spectrophotometric measurements.
The FDA stipulates that at least 25% of subjects must fall into each of three groups: those with light, medium, and dark skin tones, and that half of the participants in the dark-skinned group must also include individuals with very dark skin tones. includes skin color. The guidelines also stated that if equal performance is demonstrated, the device package insert should include a prominent statement stating: “This pulse oximeter has been evaluated to perform comparably between groups of individuals with a wide variety of skin tones.”
The new guidelines come decades after research showed that commonly used devices work less well in patients with darker skin tones. new research into the problem was published during the Covid-19 pandemic, when the devices played an important role. Subsequent studies have shown that the problem has led to some patients not being admitted to emergency rooms due to severe cases of Covid not receiving supplemental oxygen they needed during their intensive care treatment, and may have led to a higher number of cases organ failure and death for patients with dark skin.
“So many people have been waiting for this. I think it’s definitely a great development and a good step in the right direction,” Sara Gerke, associate professor at the University of Illinois College of Law, told STAT. “The FDA is now very clearly addressing the issue of racial bias.”
Last guidelines released in 2013 recommended that the devices be tested on “at least two darkly pigmented” or 15% of all subjects, whichever is greater, and did not specify how to measure skin color, which critics said allowed device makers to fail to target the darkest skin to take with you where equipment may not work as well.
The guidance comes after a Safety communication 2021 from the agency on the devices, days of advisory board meetings, a detailed white paper, and extensive studies commissioned by the FDA in a specialized laboratory.
“Our draft recommendations are based on the best available science to address concerns about the varying performance of pulse oximeters based on an individual’s skin pigmentation,” said Michelle Tarver, director of the FDA’s Center for Devices and Radiological Health, in a statement.
The new guidelines are not legally enforceable, but “describe the Agency’s current thinking on a topic and should be viewed as recommendations only,” the document said. Such guidelines are generally followed by manufacturers seeking regulatory approval to bring devices to market. The agency is seeking public comment on the guidelines for the next 60 days before making them final.
Tom Valley is a doctor who is eagerly awaiting the new guidelines. Valley, an ICU physician and associate professor at the University of Michigan, co-authored a 2020 article in the New England Journal of Medicine that brought renewed attention to the issue of racial disparities in how well pulse oximeters worked. While he called the proposed guidelines “an important step,” he said he was concerned they did not go far enough.
“One reason this problem has persisted for so long is that pulse oximeters are rarely, if ever, tested on the patients whose oxygen levels we need most – sick patients, often in the hospital,” he said. While the agency suggested that manufacturers could use “real-world data” in some cases, such data is difficult and expensive to collect, and Valley noted that the draft did not provide many details about how that data might be collected. “If we want to ensure that devices work well for the patients they are intended for, we need to make sure that we actually test these devices on those patients,” he said.
The agency also raised concerns about devices that may not work equally well on all skin tones but were previously approved under the less burdensome 510(k) submission path, where manufacturers only have to prove that their devices are as safe and effective as those already on the market. The agency said it believed that some devices now in use may require “significant hardware or software modifications” to demonstrate that they work well on all skin tones, but that others would likely meet the new guidelines without such changes.
The agency called on manufacturers to collect clinical data demonstrating the efficacy of these devices on all skin tones and submit that data and a new label showing that the device performs well on all skin tones in a new 510(k) submission, which would be reviewed in 30 days. The agency said it would maintain a list on its website of devices that have been shown to work equally well on all skin tones “to further promote transparency.”
Carmel Shachar, an assistant professor of clinical law at Harvard School of Law, said the public website would be a carrot and not a stick for the FDA to use with manufacturers. “You have a label that says, ‘My device is not biased.’ You have this website where, ideally, physicians and anyone in the hospital who is responsible for ordering devices will look for devices with this label, go to the list of devices and choose from those devices,” she said.
Valley said such information would also be useful for doctors. “What we all want is to know which pulse oximeters to use every day,” he said.
The draft guidance applies to devices used in hospitals and doctor’s offices and does not apply to devices sold directly to consumers for wellness or athletic purposes that are not subject to FDA review. Gerke said she believed these over-the-counter devices should be subject to the same standards to prove they work equally well for everyone, since they are often used by consumers for medical reasons.
Lawmakers have been pushing the FDA to take action on this issue since 2021, and a handful of pulse oximeter makers, including Medtronic, have settled a lawsuit against them by an Oakland, California, clinic that serves a large Black population.
In a statement about the new guidelines, Medtronic officials said they are “deeply committed to achieving equitable care for all patients” and said the company was already testing devices using the Monk Skin Tone scale to to ensure equality and intended to comment to the FDA.
Many who rely on the devices remain frustrated by the time the agency has spent on the issue and doubt whether voluntary guidance will encourage manufacturers to improve the accuracy of oximeters without enforcement.
Theodore Iwashyna, a critical care physician and professor at Johns Hopkins University, co-authored a JAMA paper released last week that found little action was taken by device makers after the agency’s 2013 guidelines about the devices. “Without enforcement and compliance,” he said Monday, “I’m not sure why we would expect these ‘proposed new draft recommendations’ to result in better products, and therefore better, more equitable care.”
This story has been updated with a comment from Medtronic.
STAT’s coverage of health inequalities is supported by a grant from the Commonwealth Fund. Our financial supporters are not involved in decisions about our journalism.