What Is Normal Oxygen Saturation by Age?

Normal oxygen saturation for a healthy adult at sea level typically falls between 95% and 99%, but that number shifts depending on your age, your altitude, whether you are awake or asleep, and even how old the lungs doing the work happen to be. Newborns start life with saturation levels far lower than you might expect, children gradually climb to a peak, and older adults drift slightly downward as the respiratory system ages. The story is richer than a single “normal” number can capture.

The First Minutes of Life

If you have ever watched a pulse oximeter clip onto a newborn’s foot in the delivery room, you might have been alarmed by what you saw. At one minute after birth, the median oxygen saturation is only around 66–71%, depending on the study. That is not an emergency; it is expected. In the womb, a fetus relies on the placenta for oxygen, and the lungs are filled with fluid. Once the baby takes its first breaths, it takes several minutes for oxygen saturation to climb.

A large study of 468 healthy infants found that the 50th percentile for oxygen saturation was 66% at one minute, 73% at two minutes, and 89% at five minutes. It took a median of about eight minutes to reach a level above 90%.¹ A study of 130 vigorous full-term neonates reported similar figures: a median of 67% at one minute and 89% by the fourth minute, with values reaching above 90% on average by the fifth minute.²

The mode of delivery matters, too. Babies born vaginally tend to reach higher saturations faster than those delivered by cesarean section. One study measured median values of 71%, 92%, and 98% at one, five, and ten minutes respectively for vaginal deliveries, compared with 70%, 79%, and 96% at the same time points for cesarean births.³ The difference likely reflects differences in fluid clearance from the lungs and the timing of the first sustained breaths.

Infants and Children

Once the transition from fetal circulation is complete, healthy infants settle into a saturation range that looks much closer to adult values. Overnight monitoring of term infants found median nocturnal oxygen saturation between 98% and 99% over the first two years of life.⁴ That said, these numbers come from sea-level populations. At higher elevations, things look different, and altitude effects are especially pronounced in the youngest age groups.

A study of infants living at about 1,610 meters (roughly the altitude of Denver, Colorado) found the mean oxygen saturation at 24 to 48 hours of age was 92–93%. The lower boundary of the normal range at that altitude dipped as low as 85% during feeding and 86% during quiet sleep.⁵ These numbers would be worrying at sea level but are typical for that elevation.

Among children at altitude more broadly, a study of healthy kids living at moderate-to-high altitude in Papua New Guinea found a median saturation of 95%, with a normal range spanning 89–99%. Younger children, those with parents who smoked, and those who were asleep had lower readings.⁶ A systematic review of childhood oxygen saturation at high altitude confirmed that readings improve with age through childhood, that sleep lowers saturation compared with wakefulness, and that the gap between awake and asleep readings is most pronounced in the first months of life.⁷

Healthy Adults at Sea Level

For adults breathing room air at or near sea level, a resting oxygen saturation of 95–99% is the widely accepted normal range. Clinical guidelines generally consider anything below 94% worth investigating and anything at or below 90% a sign of clinically meaningful hypoxemia. Society guidelines recommend targeting at least above 90%, with a narrower window of 88–92% for patients who retain carbon dioxide.⁸

Healthy pregnant women maintain saturations within the normal adult range. A study that measured saturation during exercise in healthy pregnancies found a mean drop of just 0.3% with exertion, and no participant fell below 95%.⁹ Pregnancy increases blood volume and cardiac output, which generally keeps oxygen delivery robust despite the added metabolic demand.

How Aging Changes the Numbers

A gradual decline in oxygen saturation is a normal part of getting older. The respiratory system matures by about age 20–25, and from that point forward lung function progressively declines.¹⁰ Several structural changes drive this. The alveoli, the tiny air sacs where gas exchange happens, dilate and lose surface area. The chest wall stiffens, making each breath require more work. Respiratory muscle strength decreases. And the matching between ventilation and blood flow within the lungs becomes less even, creating zones where air and blood are not meeting as efficiently as they once did.¹¹

Despite all that, the respiratory system remains capable of adequate gas exchange at rest and during exertion throughout the lifespan. The result is only a slight decrease in arterial oxygen levels, with no meaningful change in carbon dioxide tension.¹¹ In practical terms, a healthy 70-year-old at sea level might sit at 94–97% rather than 96–99%, and that small shift does not by itself signal disease.

A population-level study found that low resting oxygen saturation was associated with higher all-cause mortality, but once lung function (measured by spirometry) was accounted for, the general mortality link disappeared. The association between low saturation and death from lung disease, however, persisted even after adjusting for lung function.¹² In other words, a slightly lower saturation is expected with age, but a sharply lower reading often points to underlying lung disease rather than aging alone. The distinction matters because the former is a monitoring finding while the latter warrants clinical follow-up.

What Happens During Sleep

Everyone’s oxygen saturation dips during sleep. Breathing rate slows, muscle tone in the airway decreases, and gas exchange becomes less efficient, especially during certain sleep stages. In healthy adults, a study of patients without sleep disorders found a mean lowest overnight saturation of about 90%, with the group average during sleep sitting near 96–97%.¹³

Age amplifies the overnight dip. In that same study, adults over 60 had lower sleep saturation values than younger participants. A separate comparison found that during sleep, half of older subjects dropped below 95% while none of the younger subjects did, and the older group experienced more frequent desaturation episodes.¹⁴ Brief dips are expected and generally harmless, but frequent or deep drops can be a sign of sleep-disordered breathing like obstructive sleep apnea, which becomes more common with age.

For infants, the same sleep-wake pattern holds but is more pronounced. Young infants show a wider gap between awake and asleep oxygen saturation than older children, and this gap narrows as the child grows.⁷ Healthy term infants monitored overnight maintained median nocturnal saturations of 98–99%, but individual readings could briefly dip lower without signaling a problem.⁴

Altitude Changes Everything

At sea level, the air contains plenty of oxygen for the lungs to work with. At higher elevations, the air is thinner and each breath delivers less oxygen. This pushes saturation readings downward for everyone, regardless of age or health status. A large cross-sectional study of over 6,000 healthy people from sea level up to the highest permanent human settlements in the Andes found that oxygen saturation dropped substantially with altitude, particularly above 2,500 meters.¹⁵

For children, a systematic review determined that an oxygen saturation of 90% represents the 2.5th percentile for healthy kids living at roughly 2,500 meters. That threshold drops to about 85% at 3,200 meters.¹⁶ Using sea-level cutoffs to diagnose hypoxemia in these populations would flag a large portion of perfectly healthy individuals.

Research among high-altitude native Tibetans living at 3,800–4,200 meters revealed an interesting age pattern. Infants under one year had oxygen saturation about 5–6 percentage points lower than older children, and the peak average saturation of roughly 90% was reached by about age 11. During adulthood, men showed a slight decline starting in their 20s, while women maintained peak levels through their 40s before declining in their 50s. The researchers attributed the female advantage to factors related to the reproductive years.¹⁷ The practical takeaway is that “normal” oxygen saturation at altitude is not just lower overall; it follows a different developmental curve across the lifespan.

Pulse Oximeters Are Not Perfect

Most people encounter oxygen saturation through a pulse oximeter, the small clip placed on a fingertip. These devices estimate saturation by shining light through the skin and measuring how much is absorbed by oxygenated versus deoxygenated blood. They are remarkably useful for a noninvasive tool, but they have real limitations worth understanding.

One well-documented issue is accuracy across different skin tones. A systematic review of 44 studies covering hundreds of thousands of paired measurements found that the majority of studies reported an overestimation of true arterial saturation in people with darker skin tones.¹⁸ A large study of pulse oximeters used in the NHS found that, for any given true arterial saturation, readings were on average 0.6–1.5 percentage points higher for patients with darker skin than for those with lighter skin. The practical consequence is that the false-negative rate for detecting low oxygen was substantially higher in darker-skinned patients, meaning the oximeter was more likely to display a reassuring number when the actual saturation was too low.¹⁹

Skin pigmentation is not the only factor that can throw off a reading. A literature review of pulse oximeter accuracy identified several variables that can shift readings in one direction or another:

• Overestimation: Darker skin pigmentation, certain abnormal hemoglobin types, and high skin temperature can push readings artificially upward.
• Underestimation: Low blood flow to the fingers (from cold hands, low blood pressure, or certain medications), nail polish, tattoos over the sensor site, anemia, and true hypoxemia itself can drag readings below the real value.
• Signal distortion: Motion, bright ambient light, very fast heart rates, electromagnetic interference, and poor sensor placement can scramble the signal entirely.

These factors were identified in a comprehensive review.²⁰ Motion is a particularly common problem. A volunteer study comparing three clinical-grade pulse oximeters found that accuracy during movement varied enormously between devices, with sensitivity for detecting low saturation during motion ranging from as low as 15% to as high as 100% depending on the brand.²¹

Pulse Oximetry Versus an Arterial Blood Draw

When precision truly matters, the gold standard for measuring oxygen saturation is an arterial blood gas (ABG) test, which draws blood directly from an artery and analyzes it in a lab machine. Pulse oximetry tracks closely with ABG results under normal conditions, but the agreement weakens when saturation drops below about 90%. A study in ICU patients found that pulse oximetry had high accuracy when readings were at or above 90%, but was not a reliable substitute for ABG when readings fell below that threshold.²²

A multicentre study across Australian and New Zealand hospitals found that pulse oximeters had an average bias of about -1.2 percentage points compared to ABG, meaning they tended to read slightly lower. The limits of agreement spanned roughly 6 percentage points. Reassuringly, no patient with a pulse oximeter reading of 92% or above turned out to have a true arterial saturation below 90%.²³ That 92% threshold is a useful mental benchmark: if your finger oximeter reads at or above it, the odds of actually being hypoxemic are extremely low.

The Hemoglobin-Oxygen Relationship

The reason oxygen saturation behaves the way it does, rising and falling in a particular pattern rather than linearly, is the shape of the hemoglobin-oxygen dissociation curve. Hemoglobin, the protein in red blood cells that carries oxygen, binds oxygen molecules cooperatively: once one oxygen molecule attaches, the next ones bind more easily. The result is a sigmoidal (S-shaped) curve where saturation stays reassuringly high across a wide range of oxygen levels in the blood and then drops sharply once a critical point is reached.²⁴

This is why a reading of 94% and a reading of 88% feel very different clinically, even though the gap is only six percentage points. On the steep part of that curve, a small further drop in the oxygen available to your lungs produces a large plunge in saturation. A person at 94% has a comfortable margin before anything dangerous happens. A person at 88% is already on the precipice where each breath lost matters much more. Body temperature, blood acidity, and carbon dioxide levels all shift this curve left or right, changing the oxygen level at which hemoglobin starts letting go of its cargo. Fever, for instance, shifts it rightward, which encourages oxygen unloading at the tissues but can lower the saturation number your oximeter displays.

Oxygen in Exercise

During moderate exercise, healthy people generally maintain or even slightly increase their oxygen saturation. The body responds to exertion by breathing faster and deeper, and cardiac output rises, keeping up with the muscles’ demand. Where things get interesting is at very high intensities. Elite endurance athletes sometimes exhibit exercise-induced arterial hypoxemia, in which saturation genuinely drops below 95% during maximal effort. This occurs because blood passes through the lungs too quickly for full oxygen loading, and the massive cardiac output of a well-trained heart can outstrip the lungs’ ability to oxygenate it. For most recreational exercisers, though, saturation during a hard workout stays well within normal limits.

The brain’s response to exercise adds another dimension. Research shows that during low-to-moderate intensity activity, cerebral blood flow increases by roughly 10–30% to support heightened neuronal activity. Beyond about 60–70% of maximal effort, however, blood flow to the front of the brain starts dropping back toward resting levels, largely because heavy breathing blows off carbon dioxide and causes cerebral blood vessels to constrict. Older adults have lower cerebral perfusion both at rest and during exercise, and about half of that age-related gap appears to be driven by the same carbon dioxide mechanism.²⁵ This does not change the number you see on a finger oximeter, but it means that what gets to the brain is not perfectly captured by what is measured at the fingertip.

When Consumer Oximeters Give Misleading Confidence

The COVID-19 pandemic turned fingertip pulse oximeters into household items. Millions of people now check their oxygen saturation at home, which is broadly a good thing but comes with caveats. Consumer-grade devices sold in pharmacies are not held to the same accuracy standards as hospital monitors. Readings can bounce around by two to three percentage points depending on hand temperature, finger position, and ambient light. If you see a reading of 93% on a cold morning, it may well be 95% or even 96% in reality.

The skin-tone accuracy gap is especially concerning in the home-monitoring context. During the pandemic, patients with darker skin tones were more likely to receive falsely reassuring readings, potentially delaying hospital visits. A broad review confirmed that most of the evidence points to decreased accuracy in patients with dark skin, at a level that requires particular attention.²⁶ If you have darker skin and are monitoring your saturation at home during an illness, treating a reading in the low-to-mid 90s with extra caution is reasonable.

For anyone using a home oximeter, a few simple steps improve reliability: warm your hands before measuring, sit still for at least 30 seconds, remove nail polish or artificial nails from the test finger, and take several readings a minute apart rather than trusting a single number. If multiple readings consistently sit below 94% and you feel short of breath, that warrants a call to your doctor whether or not you are older, younger, or somewhere in between.

How Pulse Oximetry Came to Exist

Continuous, noninvasive oxygen monitoring is so routine in modern medicine that it is easy to forget how recent an invention it is. In 1972, Japanese engineer Takuo Aoyagi was working on a completely different problem: trying to measure cardiac output noninvasively using a dye-dilution technique with an ear sensor. The pulsing of arterial blood kept ruining his dye measurements by introducing noise into the signal. His breakthrough came when he realized that the “noise” itself could be turned into the signal. By comparing the absorption of two wavelengths of light during the pulsatile phase of blood flow, he could calculate the ratio of oxygenated to deoxygenated hemoglobin without drawing any blood.²⁷ That insight became the foundation of every pulse oximeter in every hospital and pharmacy today. The device’s simplicity is what makes age-specific oxygen saturation data so abundant: unlike arterial blood draws, you can clip an oximeter onto a sleeping newborn or a fidgeting toddler and get a useful reading.