Pulse Oximeter Target Ranges by Age and Health Condition

For most healthy adults at sea level, a pulse oximeter should read between 95% and 100%, and in acutely ill patients receiving oxygen therapy, clinical guidelines generally target 94–98%. Those numbers shift substantially for specific conditions, age groups, and altitudes, which is why a single “normal” oximeter reading does not exist. Understanding the right target for your situation matters because both too little and too much oxygen can cause harm.

What a Pulse Oximeter Actually Measures

A pulse oximeter clips onto your finger and shines two wavelengths of light through the tissue: red light at 660 nanometers and near-infrared light at 940 nanometers. Oxygenated hemoglobin absorbs more infrared light and lets more red light pass through, while deoxygenated hemoglobin does the opposite. The device compares how much of each wavelength reaches a detector on the other side of your finger and uses that ratio to estimate the percentage of hemoglobin carrying oxygen, displayed as SpO₂.¹ That number is an estimate. The gold standard is an arterial blood gas (ABG), which measures oxygen saturation directly from a blood sample. In patients whose SpO₂ reads 90% or above, pulse oximeters track ABG values closely and can substitute for a blood draw in many situations. Below 90%, the gap between SpO₂ and the true arterial value widens enough that clinicians rely on ABG instead.²

The Standard Target for Acutely Ill Adults

The British Thoracic Society guideline, one of the most widely referenced documents on clinical oxygen use, recommends a target SpO₂ of 94–98% for acutely ill patients who are not at risk of carbon dioxide retention.³ That range is deliberately narrow. It is high enough to ensure tissues get the oxygen they need, but not so high that clinicians keep cranking up supplemental oxygen past the point of benefit. For a healthy person sitting at home, readings of 95–100% are typical and do not require any action. The 94–98% window becomes relevant when oxygen is being administered and someone has to decide when to turn it up or dial it back.

When the Target Drops to 88–92%

People with chronic obstructive pulmonary disease (COPD) and certain other chronic lung conditions live with a different oxygen equation. Their bodies have adapted to lower oxygen levels, and their breathing is partly driven by carbon dioxide buildup. Flooding these patients with supplemental oxygen can suppress their drive to breathe and cause dangerous carbon dioxide retention, a phenomenon called oxygen-induced hypercapnia.⁴ For this reason, guidelines set a lower target of 88–92% during acute flare-ups of COPD and similar conditions.⁵ If you have COPD and use a home oximeter, seeing a reading of 89% might feel alarming compared to the 95–100% range you see quoted everywhere, but for you that could be exactly where your care team wants it. The key is knowing your individual baseline and the range your doctor has specified.

Extremely Premature Infants and the Oxygen Tightrope

Neonatal oxygen targets are the starkest example of how much context matters. In extremely premature infants, the stakes of getting the target wrong are severe on both sides. A landmark trial published in the New England Journal of Medicine found that targeting SpO₂ below 90% was associated with an increased risk of death, while targeting a higher range increased the rate of retinopathy of prematurity, a condition that can damage the developing eyes.⁶ A 2026 modeling study using data from over 19,000 simulated infants estimated that universal adoption of the higher SpO₂ target range would be associated with roughly 638 fewer deaths among extremely preterm infants born in the US, but also about 551 additional cases of retinopathy requiring treatment.⁷ Neonatal units typically target SpO₂ between 90% and 95% for these babies, though the exact range and how aggressively nurses adjust alarms remain active areas of clinical debate. The core lesson: a number that is “normal” for a healthy adult could be dangerously high for a premature infant’s eyes, or dangerously low for their survival.

Pediatric Targets Beyond the Newborn Period

For older infants and children, the conversation is a bit less fraught than for neonates but still does not mirror adult guidelines. A double-blind randomized trial of infants hospitalized with bronchiolitis, the most common reason young children end up in hospital for breathing problems, found that managing to a target of 90% or higher was as safe and clinically effective as targeting 94% or higher.⁸ That result has practical implications: using a lower threshold can shorten hospital stays and reduce unnecessary oxygen use, since many children linger in hospital solely because their reading dips to 93% on a monitor. Pediatric guidelines have been gradually shifting to accept lower SpO₂ thresholds in otherwise stable children with common respiratory illnesses, though practice varies widely between hospitals.

Stroke, Heart Attack, and the Case Against Routine Extra Oxygen

It might seem intuitive that giving more oxygen to someone having a stroke or heart attack would protect their brain and heart tissue. The evidence says otherwise. A quasi-randomized trial of stroke patients found that supplemental oxygen should not routinely be given to patients who are not already hypoxic and have mild or moderate strokes.⁹ At the same time, letting oxygen levels drop too low is clearly harmful: a retrospective study of acute ischemic stroke patients found that lower oxygenation levels in the early period were strongly correlated with higher mortality.¹⁰ The practical takeaway for emergency medicine is to correct hypoxia when it exists but not to push SpO₂ higher than the standard 94–98% range in patients who are already adequately oxygenated.

Why Pushing Oxygen Too High Can Cause Harm

The risks of insufficient oxygen are obvious, but the dangers of too much oxygen are less widely appreciated. When tissues are exposed to excess oxygen for prolonged periods, the body generates reactive oxygen species that damage cell membranes, proteins, and DNA. The lungs are particularly vulnerable because they are the first tissue exposed to inhaled oxygen at high concentrations.¹¹ This is not a theoretical concern. A large systematic review and meta-analysis covering 25 randomized trials and over 16,000 acutely ill adults found that liberal oxygen therapy, defined as maintaining a median SpO₂ of about 96%, increased in-hospital mortality compared to a more conservative approach. The authors concluded that supplemental oxygen might become unfavorable above an SpO₂ range of 94–96%.¹²

More recent meta-analyses have reached slightly different conclusions on the size of the effect. A 2025 systematic review of 19 randomized trials in mechanically ventilated patients found no significant difference in mortality between liberal and conservative oxygen strategies.¹³ Another meta-analysis of five large ICU trials also found no clear mortality difference, though liberal targets were associated with slightly fewer ventilator-free days.¹⁴ The overall picture is that pushing SpO₂ well above 96% in sick patients does not help and may hurt, but the exact threshold at which harm begins is still being refined. The clinical consensus has shifted firmly away from the old “more oxygen is always better” mindset.

Altitude Changes Everything

Oximeter targets assume you are near sea level. Climb to a mountain city and the goalposts move. A large cross-sectional study measuring SpO₂ in over 6,200 healthy people at altitudes ranging from sea level to the highest permanent human settlements in the Andes found that SpO₂ decreased steadily with altitude, especially above 2,500 meters (roughly 8,200 feet).¹⁵ Even at moderate elevations, the effect is measurable. A study of healthy young adults found that resting nadir SpO₂ was 95% at moderate altitude compared to 97% at sea level.¹⁶ If you live in a place like Denver, Bogotá, or Addis Ababa, your baseline SpO₂ may sit a couple of points lower than someone at sea level, and that is physiologically normal. Applying a rigid sea-level cutoff of 95% to people living at altitude would flag many perfectly healthy individuals as hypoxic.

When Your Oximeter Might Be Lying

Even a perfectly calibrated pulse oximeter can give misleading readings under certain conditions. The device depends on detecting pulsing blood flow through the fingertip, and anything that interferes with that signal degrades accuracy.

Cold fingers: Vasoconstriction from cold temperatures significantly reduces the quality of the light signal passing through the finger, making SpO₂ estimates less reliable.¹⁷
Poor perfusion: Conditions that reduce blood flow to the extremities, including shock, severe dehydration, and peripheral vascular disease, are among the most significant limitations of pulse oximetry.¹⁸
Skin pigmentation: Darker skin can lead oximeters to overestimate SpO₂, meaning the true oxygen level is lower than displayed.¹⁹
Nail polish and movement: Dark nail polish, acrylic nails, and excessive motion can distort the light signal enough to throw off the reading.²⁰

Even in hospital settings with clinical-grade equipment and good perfusion, there is always some gap between what the oximeter reports and the true arterial oxygen saturation. A multicentre study comparing pulse oximeters with arterial blood gas values in Australian and New Zealand hospitals found an average bias of about negative 1.2%, meaning the oximeter tended to read slightly lower than reality, with limits of agreement spanning roughly negative 4.4% to positive 2.0%.²¹ In ICU patients, where perfusion is often compromised, a study found that even with an adequate perfusion index, about 16% of measurements still showed a difference between SpO₂ and SaO₂ of more than 2 percentage points.²² The takeaway is that a single reading should not prompt panic or complacency. Trends over time matter more than any individual number, and a reading that does not match how a patient looks clinically should always be investigated.

Consumer Devices Versus Medical-Grade Equipment

The COVID-19 pandemic sent millions of people to pharmacies and online retailers to buy fingertip pulse oximeters, and many now rely on smartwatch SpO₂ readings for peace of mind. These consumer devices are not useless, but their accuracy varies. A study comparing two inexpensive consumer fingertip oximeters with a medical-grade device found that all three had mean differences from a reference standard of less than 2% SpO₂, though the medical-grade device had fewer outlier readings and performed better in patients with low oxygen levels.²³

Smartwatches perform less reliably. A study testing several popular models against a clinical-grade fingertip oximeter found meaningful differences between brands. The best-performing smartwatch had a mean absolute error of about 2.2%, while the worst-performing device had a mean absolute error of about 5.8%.²⁴ An error of nearly 6% means a true SpO₂ of 92% could show up as 98% on your wrist, potentially masking a problem that warrants medical attention. In an ICU validation study, none of the tested consumer oximeters met the accepted diagnostic accuracy threshold of a root-mean-square error of 3% or less, though a hospital-grade continuous monitor came close.²⁵ For home use, an FDA-cleared fingertip oximeter is a reasonable purchase. A smartwatch SpO₂ feature is better treated as a rough indicator, not a clinical measurement.

Home Monitoring and When to Seek Help

Home pulse oximetry became a widespread practice during the COVID-19 pandemic, when hospitals set up remote monitoring programs to keep track of infected patients who were not sick enough to be admitted. Most of these programs used an SpO₂ of 92% or lower as the threshold for escalating care, such as calling the patient in for an emergency assessment.²⁶ A multisite retrospective study of over 900 patients enrolled in English home oximetry programs found that home monitoring was a safe pathway, though patients with saturations below 95% at enrollment and those who were older or had more chronic conditions were at higher risk of deterioration.²⁷

If you are using an oximeter at home to monitor a respiratory illness, a few practical rules help. Take the reading while sitting still and after the device has been on your finger for at least 30 seconds. Make sure your hands are warm and free of nail polish. Watch for trends rather than reacting to a single dip: one reading of 93% after you just climbed stairs is very different from readings consistently sitting at 93% while you are at rest. Most clinicians advise seeking medical evaluation if your resting SpO₂ is consistently below 92–94%, though your doctor may give you a different threshold based on your baseline health.

Individual Variation and the SpO₂ “Switch”

One of the less intuitive facts about oxygen saturation is that two people at the same SpO₂ can be in very different physiological states. Recent work has framed this in terms of an “SpO₂ switch,” a concept describing how individual bodies have different thresholds at which compensatory responses kick in, including changes in breathing rate, heart rate, and blood vessel behavior.²⁸ An endurance athlete acclimated to high altitude may tolerate an SpO₂ of 88% during exercise with minimal distress, while a sedentary person at sea level with the same reading might be gasping and confused. This variability is why rigid universal cutoffs can mislead. The numbers in clinical guidelines are population-level recommendations, not personalized prescriptions. Your own baseline, medical history, and how you feel at a given reading all factor into whether a particular SpO₂ value is acceptable or alarming.

This is also why clinicians are trained never to “treat the number.” A patient who looks comfortable, is speaking in full sentences, and has good color but shows an SpO₂ of 91% gets a different response than a patient at 91% who is visibly struggling to breathe. The oximeter is a tool that adds information, but it does not replace looking at the whole person.

1
Respiratory Medicine. Pulse oximetry: Understanding its basic principles facilitates appreciation of its limitations ↩︎
2
PubMed Central. Study of Oxygen Saturation by Pulse Oximetry and Arterial Blood Gas in ICU Patients: A Descriptive Cross-sectional Study ↩︎
3
PubMed Central. British Thoracic Society Guideline for oxygen use in adults in healthcare and emergency settings ↩︎
4
PubMed. Oxygen-induced hypercapnia: physiological mechanisms and clinical implications ↩︎
5
PubMed Central. British Thoracic Society Guideline for oxygen use in adults in healthcare and emergency settings ↩︎
6
PubMed. Oxygen saturation and outcomes in preterm infants ↩︎
7
PubMed Central. Modeling Oxygen Saturation Targets, Mortality, and Retinopathy Among Extremely Preterm Infants ↩︎
8
PubMed Central. Oxygen saturation targets in infants with bronchiolitis (BIDS): a double-blind, randomised, equivalence trial ↩︎
9
PubMed. Should stroke victims routinely receive supplemental oxygen? A quasi-randomized controlled trial ↩︎
10
PubMed. Association of Early Oxygenation Levels with Mortality in Acute Ischemic Stroke – A Retrospective Cohort Study ↩︎
11
PubMed Central. Consequences of hyperoxia and the toxicity of oxygen in the lung ↩︎
12
PubMed. Mortality and morbidity in acutely ill adults treated with liberal versus conservative oxygen therapy (IOTA): a systematic review and meta-analysis ↩︎
13
PubMed Central. Conservative versus liberal oxygen therapy in mechanically ventilated patients: A systematic review with meta-analysis ↩︎
14
Academic Medicine & Surgery. Comparative Efficacy of Liberal vs. Conservative Oxygenation Strategies in Adult ICU Trial Populations: A Random-Effects Meta-Analysis ↩︎
15
Thorax. Reference values for oxygen saturation from sea level to the highest human habitation in the Andes in acclimatised persons ↩︎
16
Respiration. Effect of Moderate Elevation above Sea Level on Blood Oxygen Saturation in Healthy Young Adults ↩︎
17
Biomedical Signal Processing and Control. Analysing the effects of cold, normal, and warm digits on transmittance pulse oximetry ↩︎
18
PubMed. Accuracy of pulse oximeters in measuring oxygen saturation in patients with poor peripheral perfusion: a systematic review ↩︎
19
Revista Clínica Española (English Edition). Determining factors of pulse oximetry accuracy: a literature review ↩︎
20
Revista Clínica Española (English Edition). Determining factors of pulse oximetry accuracy: a literature review ↩︎
21
PubMed Central. A multicentre prospective observational study comparing arterial blood gas values to those obtained by pulse oximeters used in adult patients attending Australian and New Zealand hospitals ↩︎
22
PubMed Central. Facing SpO2 and SaO2 discrepancies in ICU patients: is the perfusion index helpful? ↩︎
23
PubMed Central. Portable, consumer‐grade pulse oximeters are accurate for home and medical use: Implications for use in the COVID‐19 pandemic and other resource‐limited environments ↩︎
24
PLOS Digital Health. Investigating the accuracy of blood oxygen saturation measurements in common consumer smartwatches ↩︎
25
BMJ Open. Performance of popular pulse oximeters compared with simultaneous arterial oxygen saturation or clinical-grade pulse oximetry: a cross-sectional validation study in intensive care patients ↩︎
26
The Lancet Digital Health. Effectiveness and safety of pulse oximetry in remote patient monitoring of patients at home with COVID-19: a systematic review ↩︎
27
BMJ Open. Assessing the safety of home oximetry for COVID-19: a multisite retrospective observational study ↩︎
28
PubMed Central. Reframing SpO2 tolerance as a physiological switch: implications for hypoxic adaptation and exercise regulation ↩︎