A pulse oximeter reading of 94% looks reassuring on a triage sheet. But a growing body of evidence shows that for patients with darker skin pigmentation, that same reading can mask a dangerously low true oxygen level — a phenomenon researchers call "occult hypoxemia." For a publication whose readership spans clinics and field programs serving populations with a wide range of skin tones, this is not a footnote to pulse oximeter accuracy — it is close to the center of it.
What Occult Hypoxemia Means
Pulse oximeters do not measure blood oxygen directly. They estimate arterial oxygen saturation (SaO2) by shining light through tissue and reading how much of it is absorbed, reporting the result as SpO2. Occult hypoxemia describes the gap that can open between the two: a pulse oximeter reads a reassuring SpO2 above 92% while a blood gas test, drawn at the same moment, shows true SaO2 below 88% — a level that would normally prompt supplemental oxygen or closer monitoring. The reading looks normal. The physiology underneath it is not.
The Evidence: A Threefold Difference
The finding that brought this issue into mainstream clinical attention was a December 2020 study in the New England Journal of Medicine by Sjoding and colleagues, which paired thousands of pulse oximeter readings with arterial blood gas measurements. The study found occult hypoxemia in 11.4% of Black patients, compared with 3.4% of White patients — nearly three times the rate. The pattern was consistent: pulse oximeters were more likely to overestimate oxygen saturation in patients with darker skin, and the gap widened as true saturation dropped, which is precisely the range where an accurate reading matters most.
| Measure | Black patients | White patients |
|---|---|---|
| Occult hypoxemia (normal SpO2, low true SaO2) | 11.4% | 3.4% |
| Relative frequency | ~3x higher in Black patients | |
| Direction of error | Pulse oximeter overestimates saturation, more so in darker skin | |
Why Melanin Interferes With the Reading
The mechanism sits in the device’s core design. A pulse oximeter clips onto a fingertip and passes two wavelengths of light through the tissue, calculating oxygen saturation from how much each wavelength is absorbed by oxygenated versus deoxygenated blood. Melanin in the skin absorbs some of that same light. In patients with more heavily pigmented skin, this additional absorption can distort the ratio the device relies on, and the resulting error is not random noise — it skews specifically toward reporting saturation as higher than it actually is, and does so more at the lower end of the range where clinical decisions get made.
How Regulators and Manufacturers Are Responding
The regulatory response has moved in stages, and it is still moving. The U.S. Food and Drug Administration issued a safety communication in February 2021 acknowledging the limitations of pulse oximeter accuracy across skin pigmentation and other factors such as poor circulation or nail polish. In February 2024, the FDA’s Anesthesiology and Respiratory Therapy Devices Panel met to discuss revised standards for evaluating device performance across a range of skin pigments. In January 2025, the agency published draft guidance for manufacturers aimed at improving accuracy testing across pigmentation, building on existing recommendations that clinical validation studies include a meaningful share of darkly pigmented participants — specifically, at least two such participants or 15% of the study group, whichever is larger. Professional bodies have pushed further still: the American College of Chest Physicians, through the Critical Care Societies Collaborative, has urged the FDA to require that every pulse oximeter on the market demonstrate accurate, reliable readings across the full range of skin pigmentation before approval, not merely for a narrow reference population.
What This Means for Field and Global Health Programs
For programs operating in settings where patients overwhelmingly have moderately to darkly pigmented skin, this is not a rare edge case to flag and move past — it describes the population being served. A few practical implications follow directly from the evidence above:
- Do not treat a normal SpO2 as the end of the assessment. When a patient’s clinical presentation — labored breathing, altered mental status, cyanosis at the lips or nail beds — suggests hypoxemia despite a reassuring pulse oximeter reading, the clinical picture should carry more weight than the number.
- Ask vendors what their validation studies actually included. A device cleared against a narrow range of skin tones has a validation gap that matters directly for field deployment; procurement teams reviewing portable units should ask this question as a matter of course, not as an afterthought.
- Where blood gas testing is available, use it to confirm borderline decisions. It remains the reference standard the pulse oximeter is only ever approximating.
- Build the limitation into training, not just into procurement paperwork. Frontline staff who understand why a reading can mislead are better positioned to override it when the rest of the exam disagrees.
None of this argues against pulse oximetry as a tool — it remains fast, non-invasive, and indispensable in resource-constrained settings where blood gas analysis is often unavailable. But pulse oximeter accuracy is not a fixed property of the device; it varies with the patient in front of it. For a global health readership, treating that variability as clinically significant, rather than as a manufacturer’s footnote, is the more defensible position, and the one the evidence increasingly supports.