Manufacturers of pulse oximeters continue to stand by their accuracy, despite mounting evidence that the devices produce flawed readings in non-white patients.

It’s been known for decades that the fingertip devices, which measure the amount of oxygen in the blood by means of red and infrared light, can render faulty readings in patients with skin pigmentation. But the impact on patient care was thought to be negligible. 

Two new studies do more than confirm that pulse oximeters treat patients with dark skin differently. They also add to a growing body of evidence demonstrating that those differences contribute to racial and ethnic disparities in care.

One study of more than 3,000 patients, published in JAMA Internal Medicine, shows that, because the devices overestimate levels of oxygen saturation in Black, Asian and Hispanic patients, those patients received less supplemental oxygen than they needed to breathe during stays in the intensive care unit. 

The pulse oximetry readings were checked against direct measurement of blood oxygen levels — which is not practical to do in most clinical situations because it involves a painful, invasive procedure.

The other study, appearing in TheBMJ, showed that dodgy blood-oxygen readings are not limited to Black patients in the ICU. Authors studying a general care population found “occult hypoxemia” in more Black patients than white ones. That’s defined as having oxygen saturation level below 88% despite pulse oximeter readings of greater than 92%.

The findings, which researchers say could limit access to treatment for Black patients, are based on analysis of thousands of comparisons between fingertip versus arterial readings in the VA population. Both studies add to what has long been known about variation in device accuracy by patient race and the serious implications for health in the Black, Brown and Asian communities. 

“We think it’s very reasonable at this point to call upon purchasers and manufacturers to make changes [to the devices],” Dr. Eric Ward, co-author of an editorial accompanying the JAMA study, told Reuters.

When asked about the new data, makers of the devices largely stood by their accuracy and the way they’re tested across skin pigments.

Pulse oximeter maker Nonin Medical’s clip-on sensors “can be counted on to read accurately, regardless of skin pigmentation,” the company said in an emailed statement. The firm said that independent research has found its sensors have an overall accuracy of 1.3 Arms (a measure of what’s required by the Food and Drug Administration) across all skin pigmentations. 

Medtronic, for its part, said that its pulse oximeters meet FDA standards. It confirms their accuracy via “synchronized blood samples,” readings at each level of blood oxygen content and comparing the readings with measurements made from the blood sample. 

“This is done across a diverse sample of people with a range of skin pigmentations,” noted Frank Chan, president of the med-tech company’s patient monitoring business, within its medical surgical portfolio.

Whereas the FDA requires that at least 15% of patients enrolled in accuracy trials have dark skin pigmentation, both companies said the proportion of people of color in their testing reaches about 40%. Last year, the agency added a warning that pulse oximeter readings can be less accurate in patients with darker skin tones. This past June, it said it plans to convene a public meeting to explore the issue.

Meanwhile, the recent studies are “another reminder about the importance of having diverse voices at the table when decisions are being made,” said Joel Bervell, a fourth-year medical student whose TikTok and Instagram pages are devoted to pointing out racial disparities in medicine.

Bervell, who has met with the FDA about faulty pulse-ox readings among patients of color, added that “bias in technology when it comes to race is nothing new… It takes a recognition of the problem, as well as a dedication to actually address the problems, to lead to change.” 

While Bervell said he’s “thankful” the FDA is convening a committee to look into the issue, he isn’t getting his hopes up, “I’m skeptical that change will come about quickly, given that the problems with pulse oximeters have been known for decades,” he said.