Written by Carmen Wolfe
Spoon Feed
When researchers looked for occult hypoxemia by comparing arterial blood gas results to pulse oximetry measurements among patients who identified their race as Black or White, they found rates of occult hypoxemia to be almost three times higher for Black patients.
Why does this matter?
Pulse oximeters are a key tool used for emergency department triage and disposition planning, but accuracy of this tool varies based on skin tones. While this phenomenon has been described before, this correspondence offers additional evidence to support this fact and illustrates how this design flaw results in racial bias.
Bias here, bias there, bias bias everywhere
To assess the clinical significance of racial bias in pulse oximetry measurement, researchers looked for occult hypoxemia, defining this as an arterial saturation level of < 88% on ABG despite an oxygen saturation of 92-96% on pulse oximetry. They analyzed pairs of these measurements taken no more than ten minutes apart, and considered multiple confounding factors and adjusted for age, sex, and cardiovascular score on the SOFA. Researchers looked at two large patient cohorts; one at the University of Michigan Hospital in patients requiring supplemental oxygen, and the other was a pooled group of patients at 178 ICUs across the country. The University of Michigan cohort contained more than ten thousand paired samples, and in the adjusted analysis found occult hypoxemia in 11.4% (95% CI 7.6-15.2) of Black patients and 3.6% (95% CI 2.5-4.6) of White patients. Results exhibiting racial bias were echoed in the multi-center ICU cohort containing more than thirty-seven thousand paired samples, and the unadjusted analysis revealed occult hypoxemia in 17.0% (95% CI 12.2-23.3) of Black patients and 6.2% (95% CI 5.4-7.1) of White patients. Considering all of the crucial decisions we make based on pulse oximetry, these findings should serve as a reminder to acknowledge the limitations of technology and the racial bias that may be inherent in its design. Our heavy reliance on pulse oximetry may be missing occult hypoxemia in Black patients, and we should integrate this knowledge into our mental framework for decisions involving triage, supplemental oxygen delivery, and disposition.
Source
Racial Bias in Pulse Oximetry Measurement. N Engl J Med. 2020 Dec 17;383(25):2477-2478.
What about light skinned black patients?