Pulse oximeter bias undercounts sleep apnea severity in Black patients

Sleep studies often use a pulse oximeter to determine the severity of sleep apnea. But a new study by investigators from Mass General Brigham reports that pulse oximeters may undercount breathing events in Black patients, potentially leading to inaccurate severity classifications. Researchers found that the proportion of airflow-defined breathing events meeting a standard oxygen-desaturation criterion was about 20% lower in Black than white participants. The findings were published in the American Journal of Respiratory and Critical Care Medicine.

Sleep studies often use drops in blood oxygen, measured by pulse oximetry, to determine which breathing disturbances count toward sleep apnea severity. Our findings raise concern that pulse oximetry–based sleep apnea scoring may under detect clinically relevant events in some Black individuals, with potential implications for diagnosis and severity classification."

Ali Azarbarzin, PhD, first author, Division of Sleep and Circadian Disorders, Department of Medicine, Mass General Brigham

Research has shown that pulse oximeters may have a bias in their readings based on an individual's skin pigmentation. Pulse oximeters function by emitting light through a part of the body and measuring the light absorption. Melanin and other pigments in darker skin can absorb some of the emitted light, potentially affecting the accuracy of the reading.

Pulse oximetry racial biases have recently been documented in other clinical settings, including in patients presenting Covid-19 infection, but until now, it was unknown to what extent the bias might affect how obstructive sleep apnea is diagnosed or scored. To close this gap, the study analyzed data from two independent cohorts, the Multi-Ethnic Study of Atherosclerosis (MESA) and Sleep Heart Health Study (SHHS). Researchers compared severity scores after blood oxygen desaturation monitored by pulse oximetry against two metrics that do not rely on oxygen saturation: air flow and heart rate response.

The proportion of airflow-defined events meeting the 4% oxygen-desaturation criterion was 21% lower in Black than White participants in MESA and 20% lower in SHHS. Heart-rate-based scoring showed no comparable racial difference, suggesting that the difference may relate to the oxygen-based scoring process rather than to fewer breathing disturbances themselves. The heart-rate-based severity measure was also associated with cardiovascular events or death in Black participants in both cohorts, whereas oxygen-based severity measures were not significantly associated with these outcomes.

"We are not suggesting different diagnostic thresholds based on race; rather, we think the underlying measurements need to be accurate and reliable across all patients," said Azarbarzin.