An interview with Dr Nick Gompertz
Since 1990, research has suggested that finger-clip pulse oximeters may often not record blood oxygen levels accurately for people with darker skin pigmentation. Today, there’s a global call to arms to address this.
Here, we speak to Dr Nick Gompertz, former general practitioner and NHS doctor of more than 30 years, about racial bias in pulse oximetry, and what he believes medical device manufacturers need to do.
Why has racial bias in pulse oximetry become a global conversion?
In the pandemic, it became very clear that people with brown and black skin had a higher risk of death from Covid-19. One of the driving forces behind this may have been reliance on finger-clip pulse oximeters to assess patients’ blood oxygen levels and decide whether they were severe enough for treatment.
The problem with this was that the results may have been skewed for people with brown or black skin, and so people may have missed out on critical treatment.
Following the pandemic, it’s important that we continue this conversation, as finger-clip pulse oximeters are still relied upon globally to make clinical decisions.
What do you think needs to change?
I believe that there are two potential issues with finger-clip oximeters. Firstly, there’s the racial bias in many common finger oximeter readings. Secondly, there’s the fact that they don’t always guide accurate clinical decision making for anybody; whether being used on white or black skin.
That’s why I feel that the medical guidance for finger-clip pulse oximetry devices needs to be reviewed.
What I mean here is that under the international (ISO) standards for pulse oximeters, devices are allowed to pass with up to a 4% average error. This means that 66% of blood oxygen level readings will be within +/-4% of what they should be. One in 20 readings will be more than 8% inaccurate.
To put this into real terms, imagine that you were assessing someone with an oxygen level of 92%. If you were to take a reading that was 4-8% inaccurate, you might determine – incorrectly – that they had a healthy blood oxygen level of up to 96-100% (i.e. normal). They could then miss out on critical care.
A circumstance like this would be worrying for everyone, but could be more likely to happen to someone with brown or black skin. That’s why we need to address accuracy and racial bias in pulse oximetry now.
What can device manufacturers do to address racial bias in pulse oximetry?
Firstly, I think that all device manufacturers need to acknowledge the issue and perform large enough studies to determine whether their devices are racially inequitable. After all, current FDA statements confirm that the required testing is not robust enough to prove racial equality.
Secondly, they should consider if there is a way of correcting skin tone-related inaccuracies in their own devices. If there isn’t, they should produce clear warnings for users, to draw attention to the fact that blood oxygen levels may be over- or underestimated.
Another route forward is to look at alternative pulse oximetry sites – and we think we’ve found one.
Which alternative pulse oximetry site do you suggest using?
We believe that the inner ear canal presents a robust site for measuring pulse oximetry.
One reason for this is that the inner ear canal is a core (central) site (i.e. close to the blood supply to the brain and heart). This means that measurements taken from here could be more reliable when compared to measurements from finger, wrist, and arm-worn devices. Think about taking your temperature, for instance. You don’t measure it on your finger, but in your ear.
Another key reason is that the inner ear canal is the same colour for everyone. The skin here is also very thin, making it easier to measure from your blood vessels.
Therefore, pulse oximetry measurements from this site have the potential to address racial bias, while hopefully improving accuracy for all.
What are the benefits of recording oxygen from the ear, compared to other sites?
For one thing, the inner ear is a user-friendly site. Many of us wear earbuds or hearing aids regularly, and are happy to use an in-ear thermometer to take our temperature. Ultimately, it’s familiar in medical and everyday scenarios.
There also may be less movement within the ear when compared with other sites, like fingers, arms, and wrists. These areas are moving all the time. This means that many devices worn here move during daily activities and this causes inaccurate readings.
You’re currently developing EarMetrics®-Oximetry as a racially inclusive alternative to common finger-clip oximeters. How does it work?
EarMetrics®-Oximetry takes readings comfortably from the inner ear canal, which you can’t access easily via common contact sensors (such as those used in finger-clip devices).
It does this by using miniaturised sensors, which can be fitted into devices that people are using already in their normal, day-to-day lives, such as hearing aids and earphones, or in-ear thermometers.
The difference between EarMetrics®-Oximetry and other pulse oximetry devices isn’t just the site that it records from, but how it accesses the site, and what it records.
What do you think the impact of EarMetrics®-Oximetry could be?
For people to receive better and more accurate medical care, and to have trust in the treatment they receive, they need accurate oxygen readings that aren’t biased because of their skin colour. We are aiming for EarMetrics®-Oximetry to address these issues.
After all, no one wants to be over- or undertreated. I don’t believe that anyone wants to go into hospital because of a falsely low blood oxygen measurement, and they certainly don’t want to miss out on essential care because of overestimation.
How about the impact on healthcare professionals?
For healthcare professionals, I believe that EarMetrics®-Oximetry could give them confidence that their clinical decisions are based on more reliable metrics.
It could also be an easier measure to use. Think about young children, for instance. Under-fives don’t often stay still, which can make using finger-clip measures a challenge and often impossible!
EarMetrics®-Oximetry aims to also be used as part of a wider suite of tools, EarMetrics®-Healthcare, to record multiple biometrics, such as temperature, digital blood pressure and heart function indicators, and so on.
This means that in the future, professionals could perform all standard medical measurements with just one time-saving device.
How could addressing racial bias in pulse oximetry support global communities?
EarMetrics®-Oximetry sensors have the potential to become low-cost, low-power medical monitors by being fitted into everyday devices.
Once we have developed the entire EarMetrics®-Healthcare product suite further, we hope to partner with manufacturers to democratise medical health monitoring in lower and middle-income countries.
Finally, how could pulse oximetry manufacturers benefit from EarMetrics®-Oximetry?
Pulse oximetry manufacturers are already experts at producing oxygen monitoring devices. We believe that it will be quick and easy for them to switch out their current technology, and replace it with EarMetrics®-Oximetry.
Our licensable sensors simply provide a different way of measuring the same thing, but potentially with higher levels of accuracy. The commercial benefits of EarMetrics®-Oximetry for manufacturers could be truly extensive.
If you’d like to learn more about the potential of EarMetrics®-Oximetry for device manufacturers, check out our CSO David Browning’s latest blog: Use alternative sites for pulse oximetry – and gain a competitive edge.
You can also stay up to date with EarSwitch® by following us on LinkedIn, here.
Comments