Transcript: Racial Bias and Pulse Oximeters Part 3—Fixing Pulse Oximeters | Johns Hopkins

The following is a transcript of Public Health On Call’s Racial Bias and Pulse Oximeters Part 3—Fixing Pulse Oximeters.

Listen to episode →

View the transcripts for the rest of the series:

[Show Intro, Joshua Sharfstein: Welcome to Public Health On Call, a podcast from the Johns Hopkins Bloomberg School of Public Health, where we bring evidence, experience, and perspective to make sense of today’s leading health challenges. If you have questions or ideas for us, please send an email to PublicHealthQuestion@jhu.edu. That’s PublicHealthQuestion@jhu.edu for future podcast episodes.]

INTRO:

Hi, I’m Lindsay Smith Rogers, and welcome back to our special series exploring racial bias in pulse oximetry.

In the first two episodes of this series, we looked into the history of the pulse oximeter as we know it today—a device that has been shown for decades to misread oxygen levels in patients with darker skin tones—and the major systemic factors that have kept this important problem from being fixed.

We do recommend that you listen to the first two episodes, which are linked in the show notes, before this one. But just to recap: Pulse oximeters are tiny sensors usually taped or clipped to a finger that read a patient’s oxygen levels. These devices, which play an essential role in patient monitoring, rely on the passage of light to determine oxygen saturation. Because the design of these instruments makes them susceptible to differences in skin tone, pulse oximeters perform differently in people from different racial or ethnic groups, depending on their skin tone. Specifically, for people with darker skin, the results are much more likely to be erroneous, potentially falsely reassuring clinicians that the patient is getting enough oxygen when in fact they are not.

A note on language: These inaccurate readings are rooted in skin tone, not race. So, a light-skinned Black person may get a significantly more reliable reading than someone with more melanated skin. But because of this issue’s disproportionate impact on Black people, and racism being a factor in why the devices' failures are tolerated, in this series we often refer to the problem as racial bias or racial inequity.

COVID-19 brought the issue of racial bias in these devices into the headlines in 2020 and sparked fresh calls for change.

In this final episode of the series, Annalies Winny and Nicole Jurmo look into how engineers are working to improve the design of pulse oximeters, and how advocates from across the medical spectrum are leading efforts to keep the pressure on to improve equity in pulse oximetry and beyond. Here’s Annalies Winny.

Annalies Winny: We’ve gone over the issues with pulse oximeters and some systemic factors that have gotten in the way of ensuring that accurate, equitable devices are available to everyone.

But for patients, just knowing that this problem exists and that it’s finally getting more attention, that doesn’t change the immediate reality: that right now, there’s not a clear solution, and until there is, if you have a darker skin tone, this medical device may not work as well for you as it might for other patients with lighter skin tones.

Edward McClure: Oh, I'm just Edward McClure. I'm a retired advertising consultant living in Hoover, Alabama, suburban Birmingham. I’m a married father, and grandfather. I was diagnosed with COPD in 2013.

Annalies Winny: McClure is a Black patient with chronic obstructive pulmonary disease, or COPD. He uses a basic pulse oximeter at home and has one within easy reach at all times. He keeps one in his shoulder bag, carrying it everywhere he goes. And he keeps one on his bedside table. And we asked him about his relationship to this device. He said that every time it indicates his oxygen levels are dropping, he feels like he’s going to die.

Edward McClure: What it generates is fear, and that fear leads to panic attacks. And those panic attacks lead to more shortness of breath. I have to just stop and focus, attempt to tell myself, “you are not going to die.” That's the first thing I do, “you're not going to die.”

Annalies Winny: This distress was amplified when his daughter told him she’d read about racial bias in pulse oximeters and he realized that the device he relies on could betray him in a moment of life or death.

He joined Right2Breathe, a global advocacy and support group for COPD and asthma patients which has been looking into this issue. And in November 2023, McClure represented the organization and told his story during an FDA hearing to consider the problems with pulse oximeters. McClure thought about how if this was an issue for him…

Edward McClure: Well, that's gonna be an issue for a whole lot of people that needs to be addressed.

Annalies Winny: Because to him, seeing the problem and speaking up, it’s just part of being a person navigating society: doing what we can with the tools and information available.

Edward McClure: Well, that’s just what anyone would do. If they had that audience, that ear. I've always I've always gone by the thought process, we do not because we know not… So once you know something, you decide to do something, whatever you can, to deal with the issue. So whatever contribution I can make, now that I know this, that's what I will do.

Annalies Winny: In this episode, we’ll speak with some individuals who, like McClure, are driving conversations about health equity and pulse oximeters. But first it’s important to note that while the case of the pulse oximeter is a huge knot to untangle—with strings coming from across medicine, markets, regulation, academia, and even culture—is possible to make medicine more equitable. And it has happened.

For example, last year, the American Thoracic Society, the top society for lung doctors, issued an official statement that endorsed removing race as a factor in determining lung function.

The shift to race-neutral equations upends a centuries-old assumption that Black people have smaller lungs than white people. And equations used to measure lung function were calculated under the belief that there was a fundamental biological difference between races. But it’s now understood that smaller lung capacities in nonwhite populations are more likely linked to environmental exposures to pollution, or poor nutrition, and other risk factors, not biologically unalterable differences. Research shows using the new equation will have huge benefits, helping nonwhite patients move up transplant lists and increasing diagnoses and access to disability benefits.

And similarly, as of January 2023, a race-based algorithm—which made it disproportionately hard for Black patients to qualify for kidney transplants—was also replaced with a race-neutral formula. That means that clinicians should no longer use race as a factor to calculate kidney function.

And as a result, kidney programs are now required to assess their waiting lists and correct waiting times for any Black kidney candidates disadvantaged by having their kidney function overestimated due to use of a race-based calculation. As a result, more than 14,000 Black people have been moved up the transplant list.

These developments are encouraging. But what does change look like when it comes to pulse oximeter technology?

When news started to spread that pulse oximeters were in need of a serious makeover, engineers stepped up to see how technology fixes might make the difference.

Bisi Bell: We shouldn't just be trying to fix the problem after a device has already been created. I'm a proponent of thinking about these issues up front, so that the solutions can be incorporated in the design process at every step along the way.

Annalies Winny: That’s Muyinatu “Bisi” Bell—she also goes by Bye-see. She leads a team from Johns Hopkins University working on imaging technology that combines light and sound. Essentially, the technique she’s working on aims to filter out unwanted signals, resulting in clearer imaging and more accurate readings for a wider variety of people, including darker-skinned patients.

Others are reexamining how data from pulse oximeters are interpreted. A group from Washington University in St. Louis and another from Tufts University are developing pulse oximeters that calibrate according to individual skin tones, and also levels of blood perfusion, which tell us how much blood is circulating through the body.

Research teams are also experimenting with new ways of using light. Instead of using infrared light like most pulse oximeters do, one team of physicists and engineers from Brown University is working on creating a polarization-based oximeter to differentiate oxygenated and deoxygenated blood. That’s the same underlying technique that some sunglasses use to filter out certain light to create a sharper image.

Another group from the University of Texas at Arlington is using blue-green light, which can respond differently to oxygenated blood than the standard red light used in pulse oximeters. And yet another from the University of Toronto, led by Daniel Franklin…

Daniel Franklin: Is going into this concept of on-skin spectroscopy, where we increase the number of colors of light or wavelengths that we're using in order to perform oximetry in the first place.

Annalies Winny: Their theory is that adding additional wavelengths beyond the two used in conventional pulse oximetry will allow them to account for the presence of melanin—that’s the pigment that darkens skin. Though these approaches and several others might seem promising, they’re still in the research and early testing stages, and it will be years before any of these ideas could reach the wider market and potentially replace or be incorporated into devices that are already out there.

In the meantime, to accelerate research and improve our understanding of the devices on the market, researchers like Franklin say we need more transparency around pulse ox data. He says that projects like the Open Oximetry Lab at the University of California San Francisco, which independently evaluates pulse oximeters, are essential to achieving that. After all, fixing the physical devices themselves is just one part of the solution.

Daniel Franklin: Hopefully, those types of initiatives will then make these things in the public domain to where companies don't have to keep these trade secrets. Ultimately the hardware is not really that special, to be honest. The special sauce for these companies comes in that algorithm and the calibration factors that take that raw data and then compute an oximetry value from it. I think it's only through publishing these data sets and having them be open that you'll be able to make real progress.

The solution will come through the initiatives where you bring engineers like myself, with policy makers, with clinicians, with patients all into the same room in order to not just quantify the problem but also to create open source devices and open publications that show solutions and possible results.

Annalies Winny: Cracking open what has long been a black box of proprietary data on these devices, in a multibillion dollar pulse oximeter market, is no easy task. But those with purchasing power can make their voices heard.

Here’s Tom Valley again—he’s the pulmonary critical care physician we heard from in episode one:

Tom Valley: I would say that it's incumbent on health systems and clinicians to demand that, like, the devices that they're using, be shown to work well, regardless of the color of your skin.

And I think ultimately, that might be the only way to motivate change is to make purchasing decisions that are based on demonstrated data that these devices work well, for everyone. And that might then motivate companies to either release their data or make a change.

Annalies Winny: And just as health care systems should demand better information, high quality patient data is also critical to ensuring a future where it doesn’t take decades just to agree that there’s a problem.

John Jackson, an associate professor in epidemiology at the Johns Hopkins Bloomberg School of Public Health—we spoke to him in episode two—he thinks a lot about creating systems built with equity at the core. He believes that lack of data on race, ethnicity, and other social characteristics—both in electronic medical records and within other administrative data—makes it impossible to evaluate the performance of medical products in real world conditions.

John Jackson: I think one of the key things that we need to do is to improve the collection of racial and ethnic data and other data on social characteristics. So that we are able to actually stratify these characteristics and to evaluate the performance using these secondary data sources.

Annalies Winny: In other words, using data collected in hospitals around race and health outcomes can help find patterns and connections that clinicians might otherwise miss.

John Jackson: And it would be a major investment, but it's probably a minimal thing that we would need to be able to evaluate products in real time.

Annalies Winny: Another approach to pushing for change in pulse oximetry: taking the issue to court.

Dr. Noha Aboelata, who we met in the last episode, is another physician-turned-advocate. She’s CEO of Roots Community Health Center, a health advocacy organization in Oakland, California. A couple of years ago, the organization wrote to over a dozen pulse ox manufacturers, demanding that they warn users about the flaws with the devices. They…

Noha Aboelata: Didn't get much of a response. We got one company that agreed to affix a label to the product that they sold in California. We really made no progress with the others. And so after about a year, we decided to file the lawsuit.

Annalies Winny: So last year, they sued Walgreens, CVS, GE HealthCare, and top pulse ox manufacturers like Masimo, demanding that they halt further sales of unreliable devices, and that manufacturers and distributors be required to warn purchasers about the devices’ shortcomings. And in the process…

Noha Aboelata: We did meet with the state attorney general's office, to raise their awareness of it in hopes that they might also consider taking some kind of action. Our attorney general did—along with about two dozen other attorneys general across the United States—submit a letter to the FDA with the expectation that this be addressed.

Annalies Winny: The letter put pressure on the FDA to act faster to address the inequity, and said, quote, “it is unconscionable in a multi-racial, multi-ethnic nation … [that] the pulse oximeter is not calibrated to adjust for skin color.”

The timing was apt: After a year of inaction, shortly after the letter was sent, the FDA scheduled a second hearing to address the issue of inaccurate pulse oximeters.

Increasingly, this kind of physician advocacy is being built into the fabric of the profession. For the current generation of medical students and trainees, that can mean advocating for change on social media.

Fourth year med student at Washington State University, Joel Bervell, who is Ghanaian-American and perhaps better known as TikTok’s “medical mythbuster,” is using his platform to highlight racial bias across the medical spectrum. Bervell has amassed more than 700,000 TikTok followers, and it all started with a video he posted in December 2020 that went viral. It was about pulse oximeters.

Joel Bervell: It was winter break, so I wasn't kind of expecting anything of it. Woke up the next day and had over half a million views on the video and thousands of comments from health care professionals saying, either I didn't know this. I use it every single day, or from patients, saying, “Wow! How did I not know that this disparity existed? Is this what happened to myself or my loved one?”

And it made me realize just how little people understood about healthcare disparities that manifest in this way, and how can people be perpetuating biases without even realizing it.

Annalies Winny: TikTok’s short form videos have proven to be a really effective way to get the word out and open conversations around racial bias in pulse oximetry and beyond. Those are conversations that historically haven’t had a formal space to occur.

Joel Bervell: It creates this conversation that otherwise wouldn't exist. I think, specifically for communities of color, we've always found spaces and ways to be able to have conversations that aren't being held anywhere else. In today's day and age, that's social media.

Annalies Winny: Bervell says his TikTok videos have even been cited in medical school syllabi.

Joel Bervell: My friends will send me pictures from the syllabus all the time, and say. Oh, like this specific TikTok is there, or I saw your name here. That's because there is no central place to talk about all these disparities.

Annalies Winny: The lack of formal conversations about race and racism in medicine is a symptom of another major problem: the fact that people of color are not well represented in medicine generally.

Andrea Deyrup, who we met in episode two, is a professor of pathology at Duke University School of Medicine and an advocate for making medicine more inclusive.

Andrea Deyrup: I think the real crux of the matter is that we have so little diversity in medicine, and that those individuals who are not part of the dominant subgroup really are very challenged. They are often undermined or pushed to the side.

Annalies Winny: She says that changes to medical school curricula shouldn’t be limited to simply informing future clinicians around this general issue of racism, it’s about equipping med students with specific skills to look at what they’re taught critically, and to speak up when something doesn’t make sense or doesn’t represent their experience,

It’s not an exaggeration to say that early reports from clinicians we’ve mentioned throughout the series, including Michael Sdojing and Tom Valley, and anthropologists like Amy Moran-Thomas, are what started this cascading series of events, this reckoning, around racial bias in pulse oximetry.

While Bervell told us that he didn’t learn about pulse oximetry bias in medical school, he was taught to be an advocate. At Wash U, where he’s studying…

Joel Bervell: I think they did an incredible job of building this next generation of physicians that are not just doctors, but are advocates for their patients and for patients they may not even meet yet.

Annalies Winny: Another essential part of the solution is patients advocating for themselves and their loved ones. Michael Abrams is a senior health researcher at Public Citizen, a consumer advocacy organization , where he focuses on the safety and efficacy of drugs and medical devices. He says that the only way to really overcome the inequities in pulse oximetry is…

Michael Abrams: With a collective approach to acknowledging those harms and trying to ameliorate them. By collective, that includes not just people with power, but people who are often victimized because of these things, and them looking out for themselves with information.

That people should speak up—speak up for your loved ones who are being cared for, speak up for yourself when you're being cared for. That is exceedingly important and powerful, not only to just changing the immediacy of your treatment, but also to seeing that in the future we get better medical devices and better medical technologies that work the way that they're supposed to on everybody.

Annalies Winny: Joseph Wright, who we met in the first episode, is a pediatric emergency physician and Chief Health Equity Officer at the American Academy of Pediatrics. He constantly grapples with health inequities in his advocacy work, but often most troubling is watching them playing out in the exam room.

Joseph Wright: One of the biggest challenges that I am having is in having conversations with patients and educating them about how they or their family may have been impacted.

As an African American physician, communicating with African American patients or other patients of color, the dynamic is one of feeling, well, this is historically what happens to us, and you're just validating what we already knew, But to hear it, even though you're addressing it from the standpoint of trying to fix the problem, it is hurtful.

Annalies Winny: As we draw this podcast series to a close, permit us some final observations.

We read everything we could find for these episodes and spoke to dozens of experts. What we did not find might be as interesting as what we did find.

We did not find an evil scientist who set out to make pulse oximeters that discriminate on the basis of skin color. We also did not find a company business plan that intended to make money off of racial discrimination. And as far as we can tell, nobody set a goal for thousands and thousands of patients to be struggling to breathe and yet be told their oxygen levels were fine.

But it happened anyway. It happened because many did not pay attention to alarming reports in the scientific literature. It happened because companies rushed to market before asking essential questions. It happened because regulators didn't set reasonable standards. It happened because racial bias is like a magnet, skewing our world, one small bias after another, adding up to serious harm.

And now, even when the problem of a deadly racial bias in pulse oximeters is in plain sight, even when there seems to be broad acknowledgement that what's happening is absolutely unacceptable, the path forward is still unclear.

The FDA, companies, and the medical community all must take urgent action. And yet when difficult problems require a lot of people and organizations to fix, that fact can get used as an excuse for no one to act decisively.

Those sorts of problems—ones that require us to work together to improve our collective health—are, by definition, problems of public health. That's why you've been listening to this story on a public health podcast, and that's why we've been working so hard to tell it.

Public health problems require the leadership of public health agencies. The FDA is now fully engaged in recognizing and addressing racial bias in pulse oximeters. The agency, as soon as possible, should develop a clear, morally justifiable, and scientifically rigorous approach to ensuring that there are not dangerous disparities in how these devices function. This may require setting a deadline for better devices to come to market.

And for their part, pulse ox manufacturers need to acknowledge the severity of the issue and not resist new standards. They should be investing in the next generation of devices that work for all patients. Health care systems should be ready to spend some money to replace their old, biased pulse oximeters with new, proven, and equitable devices. The science is clear that this can be done. But it will require telling and retelling this story to get people to see the solution through to the end.

Pulse oximeters are but one example of racial bias in medicine. And success with these devices can define a path that other issues can follow. It won't be easy. But one day soon, we hope, we’ll all be able to breathe easier.

And for the last word, let’s turn back to Joseph Wright.

Joseph Wright: We are moving into an environment where our approaches to clinical care have to be race-conscious. In other words, very cognizant of the fact that bias could very well be part of the root cause of particular decisions around the use or the approval of a certain technology.

Lindsay Smith Rogers: Thanks for listening to episode three of our three-part series. Episodes one and two are linked in the show notes. Special thanks to Annalies Winny and Nicole Jurmo for co-producing this series.

[Podcast Credits, Joshua Sharfstein: Public Health On Call is a podcast from the Johns Hopkins Bloomberg School of Public Health, produced by Joshua Sharfstein, Lindsay Smith Rogers, Stephanie Desmon, and Grace Fernandez Cecere. Audio production by JB Arbogast, Holly Cardinell, Spencer Greer, Matthew Martin, and Phillip Porter, with support from Chip Hickey. Distribution by Nick Moran. Production management by Catherine Ricardo. Social media run by Grace Fernandez Cecere. Analytics by Aliza Rosen. If you have questions or ideas for us, please send us an email to PublicHealthQuestion@jhu.edu. That's PublicHealthQuestion@jhu.edu for future podcast episodes. Thank you for listening.]