FDA advisors want standards, labeling to address racial disparities with pulse oximeters

Regulatory NewsRegulatory News | 03 November 2022 |  By 

(Source: Pexels)

A panel of external US Food and Drug Administration (FDA) experts are urging more research to improve the efficacy of pulse oximeters in people with darker skin, creating a standard skin pigmentation assessment and better labeling guidelines to ensure users know the limitations of the devices.
 
On 1 November, FDA convened a meeting of its Anesthesiology and Respiratory Therapy Devices Panel to discuss how to improve the development of pulse oximeters and how they are regulated. While pulse oximeters have been on the market for decades, their use skyrocketed during the COVID-19 pandemic as people bought over-the-counter devices to monitor their oxygen levels. Their increased use drew attention to the fact that the devices were not as accurate as expected for people with darker skin.
 
The issue was first brought to light by University of Michigan (UofM) researchers, who noted in a letter to the editor published in the New England Journal of Medicine in December 2020, that when pulse oximeters were first brought to market their development did not include racially diverse populations. Based on a retrospective study, the researchers found patients who self-identified as black were three times as likely to have hypoxemia not detected by pulse oximeters when compared to white patients. Subsequent studies have revealed that there may be a link between skin pigmentation and false pulse oximeter readings that can lead to failed or delayed treatment for hypoxia.
 
“The issue of racial disparities in pulse oximetry is of great public health importance and further highlights the need to ensure that all medical devices are safe and effective in all the populations in which they are intended to be used,” said Jeff Shuren, director of FDA’s Center for Devices and Radiological Health (CDRH) at the start of the 10-hour panel meeting. He also noted that addressing such racial disparities in health care was a key element of CDRH’s 2023-2027 priorities.
 
The panel heard from a slew of researchers and experts about why pulse oximeters may be less effective in people with darker skin, which is likely due to differences in interpreting readings based on melanin levels and skin pigmentation. The meeting also focused on expectations for consumer and medical grade pulse oximeters, as well as those used under physician observation in hospital settings versus those used at home by patients.
 
The experts grappled with ways to account for skin pigmentation differences to give better readings and inform patients. Some ideas included adding more light-emitting diodes to get improved readings, better software algorithms to determine skin pigmentation and using a skin pigmentation scale to calibrate the devices.
 
Murad Alam, a dermatologist at Northwestern University, noted that intense pulse light (IPL) devices used for hair removal are designed to pick up information on people’s skin type so they can be adjusted to prevent them from burning. He said that such technology may be able to assess skin pigmentation.
 
"We might be, at least hypothetically, trying to measure something precisely that is inherently not precise," Alam said. He noted that it may be infeasible to develop instruments that vary dramatically depending on the individual, the time of day and how it's applied. Instead, he said the best way may be to look at developing large skin tone categories.
 
Like other panelists, Alam said that if trying to be precise, it may be best to develop algorithms to calibrate devices to different skin tones.
"I think the algorithmic mode makes sense and I don't think we should slice pigments too finely (because) I think we'll get into trouble," he added.
 
While much of the focus of the meeting was on devices used in the hospitals that are considered diagnostics, Alam cautioned that over-the-counter (OTC) pulse oximeters are more widely used and are used by individuals to make important medical decisions. While specialist physicians can compensate for differences in the devices, the average person using these devices at home can’t. He also warned that OTC pulse oximeters shouldn't be banned from use because that would be like "throwing out the baby with the bath water."
 
Hugh Cassiere, an internist at North Shore University Hospital, said skin pigmentation calibration on the devices should be based on self-identification because for “now that seems to work pretty well.” Eventually, he said if there are studies that develop a better scale for skin pigmentation, the agency may want to reevaluate the issue.
 
"We just do not have enough studies that point to skin color scales defining the problem that we've found,” said Cassiere. “We found the problem through self-identification, not through skin tone studies.”
 
Steven Nathan, who chairs the panel and is the medical director of Inova Health's advanced lung disease and lung transplant programs, questioned whether a scale was even necessary.
 
"As long as manufacturers can show that it works adequately and as well in people of different colorations, then it's up to the manufacturers how they get there rather than mandating use this scale or that scale," he said.
 
One of the biggest issues discussed during the meeting is FDA’s allowable error margin for pulse oximeters. The agency uses an acceptable root mean square (ARMS) of 3%, which the panelists widely agreed was too permissive. Michael O’Connors, a University of Chicago anesthesiologist, said the agency should consider narrowing that to 1.5-2%.
 
O’Connor also noted that a lot of the studies referenced by FDA looked at blood oxygen saturation levels of 92% and over, which he said doesn't make a lot of sense from a clinical perspective. Instead, he said it would have been more significant to look at data where blood oxygen saturation levels were between 75% and 92%.
 
"That's the domain where pulse oximetry will in fact inform changes in decision-making," said O’Connor.
 
Inpatient versus outpatient devices
 
Arlene Hudson, chair of anesthesiology at the Uniformed Services University said that outpatient devices are really the iceberg of pulse oximeters.
 
"I don’t think most physicians know that these are fully unregulated and untested devices, and all of that is now feeding into apps and medical decision-making," said Hudson.
 
She also noted that physicians use pulse oximeters differently than patients to monitor trends, in conjunction with other medical information.
 
“I think one thing we should potentially consider is taking those numbers off the devices that are used at home, to not give individuals that false sense of security that is exactly the same thing we use in a hospital setting," said Hudson.
 
In response, O'Connor said that FDA may need to create a third class of pulse oximeters that are meant to be used at home under the supervision of health care providers. Nathan, however, said he wasn’t sure about the feasibility of a third class of devices but did say there needs to be better delineation between consumer-grade and medical-grade devices.
 
Malvina Eydelman, director of CDRH's Office of Health Technology 1, noted that medical-grade pulse oximeters are already available for patients to use at home with a prescription.
 
Cassiere said there should be no expectation that consumer-grade pulse oximeters should perform like medical-grade devices and that it's not realistic to ask manufacturers to solely make medical-grade devices because the public is using them in ways that are unintended.
 
"You can order and provide a medical-grade oximeter at home,” he said. “We're talking about creating another category that doesn't make sense to me at this point.”
 
Connor summed up the panel’s thoughts that labeling on consumer-grade devices should explicitly state that they are not meant for medical use.
 
Sean Hennessy, an epidemiologist at the University of Pennsylvania, said that when people buy a thermometer over-the-counter they expect the results to be accurate and they will have the same expectation for pulse oximeters regardless of what's on the label.
 
"I think it sends a mixed message to allow companies to sell something and yet not have confidence in the results it puts out, particularly when it's being used for health reasons,” he said. “When these things were first sold before the COVID-19 pandemic they didn't have the use they do now. I think the environment has changed and so I think the regulatory environment should change."
 
Alam argued that OTC pulse oximeters are mislabeled because they are not health and wellness devices.
 
“They are devices that are bought by people because they want to be alerted to when their health has deteriorated to a point where they need to seek immediate medical attention,” he said.
 
Alam said that FDA needs to find a way to regulate OTC consumer-grade pulse oximeters. If the current regulatory scheme is insufficient to dramatically improve the products or FDA is unable to regulate them with enough oversight to make a difference, he argued that the agency may need to reclassify them.
 
Alam also cautioned that the regulatory oversight shouldn't put so much burden that a $10 device has to cost $1000 because that would create access issues for patients.
 
"The main thing is reclassification is essential," he added.
 
Nathan questioned whether it was feasible to require companies to produce oximeters that were of only medical grade but agreed that reclassification is needed, though it would likely increase the price of the devices.
 
"The companies that make medical grade pulse oximeters are the same companies that make the non-medical grade pulse oximeters so I don't think it's a big leap for them to up their game,” said Hudson. “I think it's more of a marketing issue than anything else. I don't think we'd be asking too much of the companies.”
 
Lonny Yarmus, a pulmonologist at Johns Hopkins Hospital, recommended that FDA include a boxed warning so patients and their physicians are aware of the limitations of the devices.
 
William Wilson, the industry representative on the panel and chief medical officer at Masimo, noted that in addition to developing new standards for at-home pulse oximeters, FDA could put out a warning notice that if patients are being monitored at home, they should use a medical grade device.
 
"Not all pulse oximeters are created equal, some are better than others, but we can hold all manufacturers to a higher standard," he said.
 
Wilson said that higher standards are something FDA has to come to an agreement on, considering that some of the OTC devices are now better than what's available in the hospitals because technology has gotten better with time. He noted that FDA does allow patients’ extremities to be warmed before using pulse oximeters which may skew data. He argued the agency shouldn't allow that practice anymore and the basic requirement should be normal room temperature that would allow comparison of apples to apples with devices.
 
Nathan summed up the committee’s thoughts and said he agreed that the available clinical evidence does demonstrate that there are performance differences between people who use pulse oximeters depending on their skin pigmentation. That difference in performance has led to increased risk for patients.
 
The panel also agreed there should be standardization of skin pigmentation assessments which should ideally include an objective scale but could also include more subjective skin tone scales such as the Fitzpatrick scale.
 
Nathan noted that self-identified race should also be included in that process. He also encouraged using the pulse oximeters on multiple sites on the body such as on the finger, ear lobe and forehead to fully evaluate the performance characteristics of the devices.
 
The panel also agreed that FDA’s ARMS threshold needs to be tightened, though to what degree should be left to the statisticians. It also agreed that there needs to be accuracy across skin pigmentations and there needs to be transparency if there are any inaccuracies. How much accuracy is permissible is yet unknown.
 
Finally, the panel agreed that at home pulse oximeters need to include labeling that clearly states they are not for medical use and contains clear and concise language and images to instruct people on proper use of the device.
 
Nathan also recommended including a clear statement on the product that states, "Not FDA approved for medical use." On prescription devices for home use, the panelists said it was up to the prescribing physician to instruct the patient on how to properly use and read the device.

 

© 2023 Regulatory Affairs Professionals Society.

Discover more of what matters to you

0;8;10;11;13;15;20;30;31;32;