A fingertip pulse oximeter is one of the cheapest, fastest ways to check whether a patient’s blood is carrying enough oxygen, which is exactly why it has become standard kit for clinics, ambulance crews, and field health teams working far from a hospital lab. But pulse oximeter accuracy is not uniform across patients or conditions, and a growing body of clinical evidence shows that the number on the screen can be quietly wrong in ways that matter most for the populations global health programs most often serve.
Understanding why a reading can mislead is not a technical footnote. In field and outreach settings, where a pulse oximeter reading may be the only objective vital sign informing a referral decision, knowing when to trust the device and when to look past it is part of using it responsibly.
How a Pulse Oximeter Estimates Oxygen Saturation
A pulse oximeter clipped to a fingertip shines two wavelengths of light through the tissue and measures how much of each is absorbed by the blood pulsing beneath the skin. Oxygenated and deoxygenated hemoglobin absorb light differently, and the device uses that ratio to calculate an estimated arterial oxygen saturation, reported as SpO2. It is an elegant, non-invasive proxy for a value that would otherwise require drawing arterial blood, but it is still an estimate built on an optical signal that has to travel through skin, tissue, and whatever else happens to be in the light path.
The Skin-Pigmentation Accuracy Gap
The most consequential limitation documented in recent years concerns skin pigmentation. The U.S. Food and Drug Administration has acknowledged that pulse oximeters can overestimate oxygen saturation in patients with darker skin, and issued a safety communication flagging the risk of inaccurate readings under certain conditions. An FDA advisory panel convened in February 2024 to discuss the problem and recommended new clinical-trial standards requiring manufacturers to test devices across a full range of skin pigmentation before approval.
The clinical stakes were made concrete by a widely cited study published in the New England Journal of Medicine in December 2020. Researchers at Michigan Medicine, led by Michael Sjoding, analyzed nearly 50,000 paired pulse oximetry and arterial blood gas measurements and found that Black patients experienced occult hypoxemia — a true low oxygen level that the pulse oximeter failed to flag — at more than three times the rate of white patients: 11.4 percent versus 3.4 percent. A separate analysis of COVID-19 patients found pulse oximeters overestimated oxygen saturation in Black patients by roughly 0.93 to 1.2 percentage points compared with white patients. For a journal audience whose readership and subject population skews heavily toward tropical and lower-income settings, where the majority of patients are not light-skinned, this is not a marginal statistical footnote — it describes the population most likely to be screened with these devices.
Other Confounders Field Teams Should Know
Skin pigmentation is the best-documented source of systematic bias, but it is far from the only variable that can push a reading off target. The table below summarizes the confounders with the strongest evidence base.
| Factor | Effect on Reading | Field Mitigation |
|---|---|---|
| Poor perfusion / cold extremities | Weak or absent pulse signal; unreliable or missing reading | Warm the hand, try an earlobe probe, or re-check after the patient rests |
| Motion artifact (shivering, tremor, finger movement) | Erroneous values, signal loss, false alarms | Hold the extremity still; repeat the reading rather than trusting one snapshot |
| Dark nail polish (blue, green, black, brown, purple) | Can underestimate SpO2, in some studies by 3–5 percentage points | Remove polish or use a different finger |
| Carbon monoxide exposure | Falsely normal or high reading; standard oximeters cannot distinguish oxyhemoglobin from carboxyhemoglobin | Treat suspected CO exposure clinically regardless of the SpO2 reading; specialized SpCO devices exist but have low sensitivity as a screening tool |
| Severe anemia, irregular heart rhythm, poor probe placement | Reduced signal quality or misleading values | Cross-check against clinical presentation, not the number alone |
What This Means for Choosing and Using Field Units
None of this argues against pulse oximetry as a field tool — it remains fast, inexpensive, and genuinely useful for trending a patient’s condition over time. What the evidence argues for is treating a single SpO2 reading as one data point rather than a verdict. A patient who looks clinically unwell, is breathing hard, or reports breathlessness warrants escalation even when the oximeter reads reassuringly, particularly in patients with darker skin where the occult-hypoxemia risk is documented and real. Programs selecting portable units for field use should also favor devices validated across a diverse range of skin tones where that information is available, since regulatory standards on this point are still catching up to the clinical evidence.
Practically, that means training field staff to check for the basics before trusting a number: is the finger warm, still, and free of dark polish; is the waveform on the device stable rather than erratic; and does the reading match what the patient looks like. A pulse oximeter that reads 96 percent on a patient gasping for breath is not a reason to stand down — it is a reason to look harder.
Conclusion
Pulse oximeter accuracy depends on more than the device sitting in the kit bag. Skin pigmentation, perfusion, motion, nail polish, and carbon monoxide exposure can all shift the number that appears on the screen, and the clinical evidence on pigmentation bias in particular has direct relevance for the patient populations most field and global health programs actually serve. Reading the device correctly means reading the patient alongside it.