Racial biases continue to be exposed in medical devices clinicians commonly use to evaluate patients. From thermometers to oximeters to X-rays, people of color receive the most inconsistent results, and it’s bad for their health.
Health systems are attempting to weed out these biases to boost quality and eliminate health disparities. Many are still in the process of identifying where inequities exist and what processes and devices should be further evaluated. The Food and Drug Administration is starting to take action, too.
Some medical devices generate racial bias in their results—such as pulse oximeters and infrared thermometers—largely because they weren’t tested on diverse groups of people prior to going to market. Researchers are now stepping back to analyze medical devices’ performance across all demographics and finding volatility in their readings of people with darker skin tones.
A recent Emory University study details inconsistencies in measuring the body temperature of Black patients using forehead thermometers that had already been evaluated in more than 100 clinical studies. There was no robust evaluation of the accuracy by race prior to its widespread adoption, researchers found.
The odds of detecting a fever in Black patients using forehead thermometers was 26% lower than if an oral thermometer were used, according to the study. That discrepancy didn’t exist for white people, indicating a disparity in the way infrared thermometers measure temperature in people with different skin tones. Missed fevers could lead to delays in diagnoses, delays in antibiotic treatments and increased death rates among Black patients.
“This is an important finding because health systems routinely use fever cut-offs to alert or notify team members about care pathways, such as sepsis alerts to deliver timely triage and antibiotics. If fevers are going undetected, then alerts are not being activated,” said Sivasubramanium Bhavani, assistant professor at the Emory University School of Medicine and author of the study.
A study by the University of Michigan Hospital in Ann Arbor found that pulse oximeters overestimate blood oxygen levels in patients with darker skin, which can delay necessary treatment.
The report, published by the New England Journal of Medicine in 2020, restarted a national conversation about the clinical impacts of medical devices that work less effectively on people of color. During the COVID-19 pandemic, health systems routinely used oximeters to triage patients and adjust supplemental oxygen levels. The readings directly impacted the level of care a patient would be provided.
The study found that pulse oximeters were three times less likely to detect abnormally low concentrations of oxygen in the blood of Black patients, putting them at greater risk.
“Oxygen is among the most frequently administered medical therapies, with a level that is commonly adjusted according to the reading on a pulse oximeter that measures patients’ oxygen saturation,” the report authors, led by associate professor Michael Sjoding, wrote. “Questions about pulse oximeter technology have been raised, given its original development in populations that were not racially diverse.”
Other medical devices may come with antiquated race-based guidelines or adjustments. Health systems have been looking to remove race-based guidelines in diagnosing and treatment, such as those to estimate kidney function or calculate natural birth risk.
Medical devices are no different. According to an article in the New England Journal of Medicine, processes around administering X-rays formerly included a race correction that caused Black patients to be exposed to higher levels of radiation.
The guidance came from the false belief that Black people have denser bones, bigger muscles and thicker skin. The result has been greater exposure to radiation for Black patients. The race correction was widely used throughout the 1960s, according to the paper, and has since been renounced. But its legacy endures in other products, and hospitals need to take careful consideration into how they are using race in medicine, the authors said.
“Fuller knowledge of the harms of simplistic race classifications can help prevent future mistakes with race adjustment,” wrote Harvard University science historians Itai Bavli and David Jones. “Such prevention is no small challenge. Racist biases are often implicit and unnoticed. By continuing to document the misuse of race in medicine, we can help protect patients from medical racism and work toward health justice.”
Health systems, trade associations and governing bodies are starting to organize.
Brigham and Women’s Hospital has been addressing racial bias in medical devices through their quality and safety team. Esteban Gershanik, the Boston-based hospital’s medical director for quality, safety, and equity, said when they learned that pulse oximeters are less effective on Black patients, the team provided education to clinicians on how to correct it in their clinical practices. They also incorporated the findings into their electronic health records system.
Those fixes are presumably short-term. Additional research around disparate readings in medical devices, and making sure there is greater representation in their testing, is still needed in order to develop better technology for all, Gershanik said.
“It’s a matter of ensuring that we evaluate these levels of potential biases throughout the system,” Gershanik said. “As greater scrutiny will come to these devices, we will also have that fall under our department of quality and safety, to understand if there’s any discrimination or bias when we look at patient safety or quality issues, or experience issues as well.”
Brigham and Women’s has partnered with the American Medical Association and the Joint Commission on a quality and safety peer network to exchange information on where they are finding discrimination or bias that affect patient safety and how to solve them.
The FDA is considering where it can play a role. The agency now requires device applications to report demographic breakdowns of their clinical trials, and is determining how to move forward on products already on the market that didn’t have to meet that requirement.
In November, the regulatory agency is convening an advisory body called the Anesthesiology and Respiratory Therapy Devices Panel for a discussion on racial bias found in pulse oximeters. The committee will look to evaluate and understand the association between skin pigmentation and oximeter accuracy.
“The FDA has been working on additional analysis of premarket data, as well as working with outside stakeholders, including manufacturers and testing laboratories, to analyze additional post market data to better understand how different factors including skin pigmentation may affect pulse oximeter accuracy,” the agency wrote in the meeting notice.