Respiratory Rate

Respiratory rate is the number of breaths you take per minute. A healthy adult at rest breathes about 12 to 20 times a minute, and wearables such as Apple Watch, Oura and WHOOP measure it while you sleep, when most people's nightly value stays within about half a breath per minute of their own usual level.

That stability is what makes the number useful. Your heart rate and HRV move with every hard session, drink and late night. Your sleeping respiratory rate mostly sits still, so when it climbs by one or two breaths a minute and stays there, something has usually changed. Often that something is an infection, and in wearable studies the rise has sometimes appeared before the first symptoms. This entry covers what the number measures, what counts as normal, how each device gets it and how accurate that is, what raises it, and how Titan reads it. The wearable data quality guide compares respiratory rate accuracy with other wearable metrics.

01What the number measures

Each breath has two parts, how often it happens and how deep it is. Respiratory rate counts only the first. Multiply it by tidal volume, the air moved per breath, and you get minute ventilation, the total volume you breathe each minute. Your brainstem adjusts both to keep blood carbon dioxide and oxygen where they belong, and rate is the part that responds most to things other than metabolism. Nicolò and colleagues' 2020 review in Sensors describes respiratory rate as rising with heat, low oxygen, pain, cognitive load and anticipatory anxiety, often when tidal volume and oxygen uptake barely change.

Sleep changes the pattern. Douglas and colleagues (1982) measured 19 healthy adults and found that breathing in every sleep stage was "more rapid and shallow than during wakefulness." Minute ventilation fell from 7.66 L/min awake to 7.18 L/min in non-REM sleep and 6.46 L/min in REM, with tidal volume in REM down to 73% of its waking level. You breathe less air asleep, in smaller, slightly quicker breaths.

02Normal respiratory rate for adults

Clinicians treat 12 to 20 breaths per minute as normal breathing at rest in adults (Nicolò et al., 2020). Population data put most people in the middle of that band. Rückert-Eheberg and colleagues (2025) derived resting respiratory rate from 5-minute ECGs in 2,224 German adults aged 39 to 88. The median was 15.8 breaths per minute, the 5th percentile 12.1 and the 95th percentile 20.1. The median rose in each age group past 60.

Source: Rückert-Eheberg et al. (2025), KORA-FF4 cohort, awake resting ECG.

Sleeping values land in the same range. Natarajan and colleagues (2021) analyzed nightly respiratory rate from 10,000 Fitbit users aged 20 to 69. The mean was 15.4 breaths per minute, and 90% of values fell between 11.8 and 19.2. Women under 50 breathed slightly faster than men of the same age, with the gap closing after 50. Women aged 20 to 24 averaged 16.7 and men 15.5. By 65 to 69, both averaged 14.8.

A high number within the normal band can still carry information over decades. Baumert and colleagues (2019) followed 2,686 older men and 406 older women after a night of polysomnography. A mean sleeping rate of 16 breaths per minute or more was independently associated with cardiovascular death, with a hazard ratio of 1.57 in the men and 2.58 in the women, and with all-cause death. That is an association in people in their late 70s and 80s. It does not make a 16 a problem for a 35-year-old, and one night is not a risk estimate.

03How Apple Watch measures sleeping respiratory rate

Apple Watch counts breaths with its accelerometer, the motion sensor that also counts steps. Apple introduced the feature with watchOS 8 in 2021 and described it this way: "Apple Watch uses the built-in accelerometer to measure respiratory rate while sleeping." Apple's 2024 paper on its sleep apnea feature explains why this works. The triaxial accelerometer captures "coarse motion of the body as well as fine movements including motion associated with breathing." Your chest and shoulders rise and fall with each breath, and a wrist resting on the bed or on your body moves slightly with them.

Two consequences follow from that design. The watch measures respiratory rate only while it is tracking your sleep, so it has to be on your wrist in bed with sleep tracking turned on. And any movement that swamps the breathing signal, such as rolling over or restless sleep, leaves gaps. Readings go to Apple Health, where Titan and other apps can read them.

Apple also uses the number in its Vitals app, which on most models builds a typical range for your overnight heart rate, respiratory rate, wrist temperature, blood oxygen and sleep duration. By default it notifies you when at least two overnight metrics fall outside that range, and it lists "medications, elevation changes, or illness" as examples of what might be involved.

04How Oura and WHOOP derive it

Rings and most other wrist devices take a different route. They read your pulse optically and infer breathing from the heart. When you inhale, vagal braking of the heart eases and heart rate rises a little. When you exhale, it falls. That rhythm is called respiratory sinus arrhythmia, and its frequency matches your breathing rate.

Oura says it "derives your respiratory rate exclusively from your nighttime PPG data," using the intervals between heartbeats from the same optical signal behind its resting heart rate and HRV. The Readiness tab shows your average from the previous night. Fitbit, whose method Natarajan and colleagues published in detail, pulls the respiratory peak out of the frequency spectrum of beat-to-beat intervals and prefers deep sleep, when the signal is cleanest. WHOOP reports the median respiratory rate over your main sleep period, "derived each night during the main sleep period via photoplethysmography," according to a paper by WHOOP's own scientists (Miller et al., 2020).

The heart-based method has a known blind spot. If your airway narrows during sleep, the heart can keep its usual rhythm while airflow drops. Oura notes that with sleep apnea "the actual air flow can be restricted" even when heart patterns look normal.

05How accurate wearable respiratory rate is

For healthy people in a lab, the optical devices come close. Berryhill and colleagues (2020) compared WHOOP with ECG and chest and abdomen inductance bands during polysomnography in 32 healthy adults. The bias was 0.1 breaths per minute and the precision error 1.0 breaths per minute. Natarajan and colleagues validated Fitbit's method in 28 people against polysomnography or a home sleep test. The mean absolute error was 0.46 breaths per minute and the root mean squared error 0.65.

Oura's published figure is weaker evidence. In a company analysis of 43 people aged 15 to 19, Oura's nightly average fell within 0.71 breaths per minute of an ECG-derived rate. That comparison used another estimate as the reference and has not appeared in a peer-reviewed journal.

Sleep-disordered breathing degrades every method. Jung and colleagues (2023) tested Samsung Galaxy Watch against a nasal airflow sensor in 195 sleep clinic patients. In patients without severe apnea, the overnight average had a root mean squared error of 0.46 breaths per minute and an accuracy of 99%. In severe apnea, error rose to 1.74 breaths per minute and accuracy fell to 79%.

Apple Watch has the thinnest record. The one direct comparison with polysomnography I could find is a conference paper with a single participant. Jaworski and Park (2023) had one healthy 31-year-old man sleep 15 nights wearing an Apple Watch Series 8 and polysomnography equipment. The watch averaged 14.1 breaths per minute against 15.6 from polysomnography, a mean absolute percentage error of about 10%. One person proves little in either direction. Apple has published validation work for its breathing disturbance feature, and I found no comparable figures for the respiratory rate number. Treat the Apple Watch trend as more reliable than any single value.

The practical reading is the same for every device. Nightly averages are good enough to follow your own trend. Absolute values differ between devices by a breath or more, so a switch from ring to watch means starting a new baseline.

06Why it stays so stable from night to night

Awake, you control your breathing. Talking, posture, attention and emotion all change it. Asleep, those inputs drop away and breathing follows chemical control and sleep stage. Oura describes the night as "an ideal time to measure respiration because your body is in a consistent state."

The data back this up. Miller and colleagues (2020) looked at 25,000 WHOOP members who had respiratory rate on 30 consecutive nights, 750,000 nights in all. Between people the values varied widely. Within a person, the average night-to-night standard deviation was 0.51 ± 0.20 breaths per minute. Natarajan's Fitbit data show the same pattern. Over 14 days, the coefficient of variation for most adults aged 20 to 24 ran from about 2% to 9%. Older adults had a wider spread, up to about 17% for women and 22% for men aged 65 to 69.

So if your usual night is 14.5 breaths per minute, a 15.0 is noise. A 16.5 that shows up two or three nights in a row is about four of your typical night-to-night steps above normal. Oura's guidance draws the line in a similar place. Variations of 1 to 2 breaths per minute are normal, and deviations of more than two deserve attention, especially when they last several nights.

07What raises sleeping respiratory rate

Illness and early infection

Infection is the best-documented cause of a sustained rise. In Natarajan's study of 3,236 people with COVID-19, 36.4% of those with symptoms and 23.7% of those without had at least one night at least 3 breaths per minute above their usual rate in the week around symptom onset. The rise peaked about two days after symptoms began, and group averages stayed above baseline for two to four weeks after onset.

Miller and colleagues built a model on WHOOP data from 271 people with COVID-like symptoms, 81 of whom tested positive. Using only changes in nightly respiratory rate, it flagged 20% of positive cases in the two days before symptoms and 80% by the third day of symptoms. That is a useful early warning for some people. It is not a test, and many infections will never show up in the number.

Fever works in the same direction. Nicolò and colleagues describe respiratory rate as the part of ventilation that responds most to heat, with a good association between breathing rate and body temperature. A higher respiratory rate alongside a raised wrist temperature is a common pattern when you are getting sick.

Altitude

Low oxygen at altitude drives breathing up. Nussbaumer-Ochsner and colleagues (2012) studied 16 mountaineers sleeping at 490 m and then at 4,559 m. On the first night at altitude, minute ventilation during sleep rose from 4.4 to 6.3 L/min, median oxygen saturation fell from 96% to 67%, and periodic breathing pushed the apnea-hypopnea index from 0.1 to 60.9 events per hour. Expect your respiratory rate to jump when you sleep high and to settle as you acclimatize, and check your blood oxygen at the same time. Periodic breathing is also the kind of irregular pattern that makes wearable estimates less accurate.

Hard training and overreaching

The link here is weaker than many recovery apps suggest. Nuuttila and colleagues (2025) put 24 recreational runners through three weeks of normal training, two weeks of overload and a recovery week, measuring nightly breathing rate with a Polar watch. Average breathing rate was 13.7 breaths per minute at every stage. The overload did not move it.

A year of Oura data from 17 elite canoe and kayak athletes points the other way. Foucaud and colleagues (2025) sorted 5,855 nights into recovery patterns. Respiratory rate explained about 15% of the difference between them, and athletes in the poorest pattern breathed two to three breaths per minute faster than those recovering best. Training duration and frequency did not predict which pattern a night fell into.

Taken together, respiratory rate does not measure training load. A rise after a hard block, especially with falling HRV and a rising resting heart rate, is a reason to look harder at overreaching, illness or poor sleep before you add more work.

Other causes

Oura lists several more: air quality, breathing problems, anxiety, menstrual cycle hormones and a warm bedroom. Intense exercise also raises respiratory rate for 20 to 40 minutes afterward, which matters only if you train close to bedtime.

08How Titan shows and uses respiratory rate

Titan reads respiratory rate from Apple Health. That means any device that writes it there, including Apple Watch, WHOOP and Oura's Apple Health sync, shows up the same way.

On the Health card and Vitals screen

The Health card on Today lists respiratory rate in breaths per minute next to HRV, resting heart rate and blood oxygen. The card shows the most recent sample in Apple Health. Tap it to open Vitals, where a period picker offers 7D, 30D, 60D and 6M.

For each period, Titan turns every day's readings into one daily value, the average of that day's samples. The latest day is the value on the last day of the period. The baseline is the median of your daily values, and a shaded band covers the 20th to 80th percentile, the middle 60% of your days. Titan draws the band only when the period has readings on at least 7 days covering at least 60% of the period. For HRV, a reading above your band shows in green. Respiratory rate has no better side, so Titan marks a reading outside your band in red whether it is above or below. The Titan app on Apple Watch shows respiratory rate on its RR tile.

The band describes your own history. It is not a clinical reference range and does not diagnose anything.

In the Sleep Quality calculation

Titan's headline Sleep score uses only duration, restorative sleep and awake time. Respiratory rate goes into a separate Sleep Quality calculation, with a weight of 0.06. Titan compares that day's respiratory rate with the median of your daily values over at least the last 30 days. The breathing part scores 1 at your median and reaches zero when the night is 22% away in either direction. With a median of 15 breaths per minute, zero is 3.3 breaths away, and a night at 16.5 scores about 0.60. The Sleep screen does not display Sleep Quality.

In Recovery

Titan's Recovery score comes from your overnight HRV and sleeping heart rate, each compared with your own measured nights. Respiratory rate is not a direct input. It plays a small role on one kind of night. If your watch recorded sleep but no HRV, Titan marks Recovery as Estimated with the reason "No HRV recorded overnight" and carries HRV forward from your last 7 measured nights. It then nudges that carried value with the night's other signals. A respiratory rate above your usual level nudges the estimate down, and a lower rate nudges it up. The weight is small, 0.05 of your typical HRV spread for each typical step of respiratory rate, and all the nudges together are capped at a quarter of that spread. This needs at least 7 measured nights with respiratory rate. The HRV entry covers how measured nights are scored.

What Titan does not do

Titan does not alert you to illness from respiratory rate. If you notice a rise along with other signs and then fall ill, add the Sickness tag in the Journal. That keeps the illness from being read as a training or lifestyle effect when you review your data later.

09Reading your own number

Compare your respiratory rate only with itself, from the same device. Look at several nights before reacting to one, because a single restless night or a loose watch can shift the value. A sustained rise of about 2 breaths per minute or more deserves a look at the rest of your data. Check HRV, sleeping heart rate, wrist temperature and blood oxygen, and think about recent travel, altitude, alcohol, a hard block or people around you who are sick. When several of these move together and you have no obvious explanation, treat it as a possible illness. Cut training stress and prioritize sleep. If you have trouble breathing, chest pain or a high fever, contact a clinician. None of these devices is a medical monitor.

10References

• Apple (2021). watchOS 8 brings new access, connectivity, and mindfulness features to Apple Watch. Apple Newsroom. https://www.apple.com/newsroom/2021/06/watchos-8-brings-new-access-connectivity-and-mindfulness-features-to-apple-watch/
• Apple (2024). Estimating breathing disturbances and sleep apnea risk from Apple Watch. https://www.apple.com/health/pdf/sleep-apnea/SleepApneaNotificationsonAppleWatchSeptember_2024.pdf
• Apple. Track your vitals on Apple Watch. Apple Watch User Guide. https://support.apple.com/guide/watch/vitals-apd15aa7ed96/watchos
• Baumert M et al. (2019). Mean nocturnal respiratory rate predicts cardiovascular and all-cause mortality in community-dwelling older men and women. European Respiratory Journal 54(1):1802175. https://doi.org/10.1183/13993003.02175-2018
• Berryhill S et al. (2020). Effect of wearables on sleep in healthy individuals, a randomized crossover trial and validation study. Journal of Clinical Sleep Medicine 16(5):775-783. https://doi.org/10.5664/jcsm.8356
• Douglas NJ et al. (1982). Respiration during sleep in normal man. Thorax 37(11):840-844. https://doi.org/10.1136/thx.37.11.840
• Foucaud A et al. (2025). Using unsupervised machine learning to characterize recovery patterns in elite canoe-kayak athletes across the Olympic training year. Frontiers in Sports and Active Living 7:1629924. https://doi.org/10.3389/fspor.2025.1629924
• Jaworski DJ, Park EJ (2023). Apple Watch sleep and physiological tracking compared to clinically validated actigraphy, ballistocardiography and polysomnography. 45th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) 1-4. https://doi.org/10.1109/EMBC40787.2023.10340725
• Jung H et al. (2023). Validating a consumer smartwatch for nocturnal respiratory rate measurements in sleep monitoring. Sensors 23(18):7976. https://doi.org/10.3390/s23187976
• Miller DJ et al. (2020). Analyzing changes in respiratory rate to predict the risk of COVID-19 infection. PLOS ONE 15(12):e0243693. https://doi.org/10.1371/journal.pone.0243693
• Natarajan A et al. (2021). Measurement of respiratory rate using wearable devices and applications to COVID-19 detection. npj Digital Medicine 4:136. https://doi.org/10.1038/s41746-021-00493-6
• Nicolò A et al. (2020). The importance of respiratory rate monitoring, from healthcare to sport and exercise. Sensors 20(21):6396. https://doi.org/10.3390/s20216396
• Nussbaumer-Ochsner Y et al. (2012). Effect of short-term acclimatization to high altitude on sleep and nocturnal breathing. Sleep 35(3):419-423. https://doi.org/10.5665/sleep.1708
• Nuuttila OP et al. (2025). Monitoring sleep and nightly recovery with wrist-worn wearables, links to training load and performance adaptations. Sensors 25(2):533. https://doi.org/10.3390/s25020533
• Oura (2020). How accurate is Oura's respiratory rate? The Pulse blog. https://ouraring.com/blog/how-accurate-is-ouras-respiratory-rate/
• Oura (2024). How Oura measures respiratory rate from your finger. The Pulse blog. https://ouraring.com/blog/respiratory-rate/
• Oura. Respiratory rate. Oura Member Care. https://support.ouraring.com/hc/en-us/articles/360025443174-Respiratory-Rate
• Rückert-Eheberg IM et al. (2025). Respiratory rate and its associations with disease and lifestyle factors in the general population, results from the KORA-FF4 study. PLOS ONE 20(3):e0318502. https://doi.org/10.1371/journal.pone.0318502