The Science of Removing Race From the Respiratory Equation

This post was written by Mike Hess, MPH, RRT, RPFT.

This year, as we recognize Black History Month, we can also recognize some long-overdue progress on the journey to better health equity. Biases and assumptions we have used for decades in two important diagnostic tools we use have recently been formally challenged, paving the way for more equitable care for all people living with lung problems.

Garbage In, Garbage Out

Spirometry data are obviously a key part of most long health journeys. As a refresher, numbers including the forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC) are compared to "normal" numbers based on the subjects height, age, and biological sex. That comparison yields the "predicted" percentages on a spirometry report. But how did we get those normal numbers in the first place?

The first efforts to tabulate these numbers actually started in the 1840s, not long after the spirometer was invented. Since these studies were done in Victorian-era London, they mostly included only White males. Things got a little more inclusive during the American Civil War, when researchers could include Black soldiers as part of their work. They found that Black soldiers often had lower spirometry values than their White counterparts. Today, we can recognize that those same Black soldiers often had lower-quality equipment, worse living conditions, and other factors that can influence overall health. Back then, it was generally believed that there was something genetic that caused lower lung function. Other studies using measurements like how much one's chest expanded during breathing showed no major difference between races, but the genetic difference theory managed to stick.

Fast forward to the early 2000s. Computerized spirometry tests now included automatic "corrections" based on race. However, different countries often used different tables of predicted values. That made it harder to do international comparisons, so a group of researchers tried to fix that. The Global Lung Initiative studied people from across the world to develop new standardized tables. It wasn't perfect, but since it was the largest set of data ever put together, it was a major improvement over previous tables. But the research didn't stop there. GLI research eventually suggested that there really were not major differences between race, especially since many countries had different definitions of different races. Finally, in 2023, the American Thoracic Society concluded that the downsides to using race-based measurements in pulmonary function testing (including spirometry) far outweighed any potential benefits and that clinicians should only use average reference equations to create tables of predicted values. Practice changes like this take time to become widespread, but we are finally seeing progress in making sure people are tested equally, regardless of how they look.

Seeing the Light

During the early days of the COVID-19 pandemic, it became clear that there were differences between the oxygen needs of lighter-skinned patients and those with darker complexions. Two separate groups of researchers, one from the University of Michigan (U-M) and one from Johns Hopkins Health System, looked at pulse oximetry readings from patients hospitalized with COVID-19. They compared the saturation readings from those skin probes with saturations values from arterial blood gas (ABG) samples taken at the same time. The U-M group found that 12% of Black patients with oximeter readings of 92-96% had ABG saturations of 88% or lower. Only 3% of White patients had that same gap in measurements. They then looked at older patient charts from 2016-17 from 178 hospitals across the country and found a similar trend. The Hopkins group included Asian and Hispanic patients as well, still finding that patients with non-White skin had inaccurate pulse oximetry readings at a far higher rate that their White counterparts. Unsurprisingly, they also found this led to delays in treatment for many.

As it turns out, there have been concerns about how accurate pulse oximetry readings are in people with darker skin tones for decades. Three different studies, in 1990, 2005, and 2007, all raised concerns about the problem. As most RTs know, pulse oximeters work by shining red and infrared light through the skin and seeing how much gets absorbed. Unfortunately, they cannot tell the difference between what gets absorbed by pigments in the skin, which can lead to these falsely-high readings.

Last month, the Food and Drug Administration finally took action to improve oximeter design. After years of investigation, discussion, and testimony, the agency released new guidance about how new oximeters should be tested and calibrated before they are sold. The recommendations include testing devices on a much larger group of people (150 compared with 10 in the old guidelines), collecting much more data (3,000 data points instead of 200), and calling for minimum percentages of people with different skin tones (based on objective measurements like the Monk Skin Tone Scale) before a device is launched.

We in health care still have a long way to go to reach true equity. However, we can take a moment to appreciate that we are at least a few steps closer to diagnosing people not on the color of their skin, but the true content of their clinical picture!