Respiratory Rate While Sleeping: The Early Signal Most Wearable Apps Ignore

TL;DR

Overnight respiratory rate is one of the oldest vital signs in medicine and among the last to show up on consumer wearables — which is partly why most apps don't feature it prominently. It reflects your breathing effort and autonomic state during sleep, and it tends to move before resting heart rate or HRV do when something in your body changes. For 90% of healthy adults, wearable data places nightly respiratory rate between 11.8 and 19.2 breaths per minute. Your personal baseline within that range, and when it shifts, is where the useful information lives.

---

A vital sign hiding in plain sight

Respiratory rate has been one of medicine's four core vital signs — alongside temperature, pulse, and blood pressure — since the 19th century. For decades it was also the one most frequently undercounted in clinical practice, because manually measuring breathing rate requires sustained, quiet observation; clinicians often estimated it or recorded a default value rather than measuring it directly.

Wrist wearables have quietly changed this. Apple Watch has tracked overnight respiratory rate since watchOS 8, reporting it in the Health app as a nightly average and incorporating it into the Vitals dashboard alongside heart rate, blood oxygen, wrist temperature, and sleep duration. Apple's Vitals feature notes that factors including medications, elevation, alcohol intake, and illness can all shift your overnight respiratory metrics outside their typical range — and that seven days of sleep tracking are needed to establish a personal baseline.

Despite this, respiratory rate remains the least-discussed metric in most wearable health conversations. Most apps lead with step counts, sleep duration, or HRV. This is partly because the absolute number — 14 or 15 breaths per minute — doesn't feel intuitively meaningful the way a resting heart rate of 48 bpm or an HRV of 60 ms might. The value of the metric comes from consistency and baseline deviation, which requires time and context to read correctly.

---

How Apple Watch estimates your breathing rate

Your Apple Watch does not contain a microphone or a pressure sensor that detects breath directly. It estimates respiratory rate using a physiological phenomenon called respiratory sinus arrhythmia (RSA): breathing naturally modulates heart rate, creating a rhythmic oscillation in the timing between heartbeats at a frequency that corresponds to your breathing rate.

Researchers at Fitbit demonstrated this method in a 2021 paper in NPJ Digital Medicine: by isolating the RSA component from the power spectral density of the heart rate interval time series, they could estimate respiratory rate with a root mean squared error of 0.65 breaths per minute and a mean absolute percentage error of 3% — in good agreement with polysomnography ground truth measurements. The same approach underlies how Apple Watch derives its respiratory rate estimate from the PPG (photoplethysmography) sensor during sleep.

The result is stored in the Health app as respiratoryRate, measured in breaths per minute, typically as a nightly average drawn from your sleep window. It is not a moment-to-moment readout; it is a summary of your overnight breathing pattern.

---

What normal looks like — and why the range is wider than you'd expect

The most commonly cited normal respiratory rate for adults is 12–20 breaths per minute at rest. But that range was defined for wakefulness, often estimated manually, and rarely comes with demographic specificity.

The Fitbit dataset in the Natarajan et al. 2021 study provides a more granular picture. Across a large, real-world sample of wearable users, 90% of healthy adult nocturnal respiratory rate values fell between 11.8 and 19.2 breaths per minute, with a mean of 15.4 breaths per minute. Several factors shifted the distribution:

• BMI: Respiratory rate reached a minimum at a BMI of approximately 25 kg/m² and rose both above and below that threshold.
• Age: Respiratory rate decreased slightly with increasing age across the adult lifespan.
• Sex: Women under 50 had higher respiratory rates than men of the same age; above 50, the difference disappeared.
• Nocturnal heart rate: Higher resting heart rate was associated with higher respiratory rate during sleep.

The practical implication is that there is no single target number that applies to everyone. What matters is your own nightly distribution — how stable it is when you're sleeping well and unstressed, and when it departs from that pattern.

---

How breathing changes across sleep stages

Respiratory rate and variability are not uniform across the night; they shift with sleep architecture.

During deep NREM sleep (N3, slow-wave sleep), breathing is most regular, steady, and metabolically efficient. The respiratory drive is consistent, tidal volume is stable, and the signal is almost metronomic compared to wakefulness. A classic study by Douglas, White, and colleagues found that minute ventilation was significantly lower across all sleep stages than during wakefulness, with the reduction greatest in REM sleep — in which tidal volume fell to 73% of the waking level.

During REM sleep, breathing becomes considerably more variable and irregular. The same cortical-level autonomic variability that makes REM sleep neurologically active also makes breathing less predictable: brief bursts of faster, shallower breathing alternate with more stable periods. This resembles the irregular breathing patterns seen in wakefulness, and it is normal.

Consumer wearables — including Apple Watch — report a single nightly average that blends these different phases. This means a night with more REM sleep may show slightly higher apparent variability in the reported number compared to a night dominated by deep NREM. It is not a flaw; it is a consequence of collapsing an architecturally dynamic signal into one number.

---

The early signal property: what shifts respiratory rate first

The "early signal" framing for respiratory rate has solid empirical support — and it's the primary reason this metric is worth tracking actively, rather than just checking occasionally.

During respiratory illness, the body's compensatory response includes both faster and less efficient breathing. The Natarajan et al. study measured nocturnal respiratory rate in a population of Fitbit users who tested positive for COVID-19. In a seven-day window spanning one day before symptom onset through five days after, 36.4% of symptomatic individuals had at least one overnight measurement 3 breaths per minute above their regular rate — and 23.7% of asymptomatic individuals showed the same pattern. The elevation often appeared before other metrics shifted, or before the person felt unwell. This doesn't mean wearable respiratory rate detects illness; it means that during illness, respiratory rate tends to be one of the first markers to shift away from baseline.

After alcohol intake, the effect is both physiologically distinct and reliably measurable. Alcohol is a muscle relaxant that reduces upper airway muscle tone during sleep, increasing the frequency and duration of breathing disruptions. A Mayo Clinic systematic review and meta-analysis of 14 randomised controlled trials found that alcohol significantly increased the apnea-hypopnea index and reduced mean blood oxygen saturation during sleep. The more disturbed breathing that results may register on your wearable as higher respiratory rate, more variation in the signal, or both — which is why Apple's Vitals app explicitly lists alcohol among the factors that push overnight metrics outside their typical range.

At altitude or elevation, lower atmospheric oxygen concentration triggers compensatory hyperventilation — breathing faster and deeper to maintain adequate oxygen delivery. This response begins quickly and typically normalises within a few nights as acclimatisation proceeds. If you travel to the mountains and notice a higher respiratory rate reading the first few nights, that is the expected physiological response, not cause for concern.

Under high training load or accumulated sleep debt, respiratory rate changes tend to be smaller than those seen during illness but directionally consistent — slightly elevated, with more night-to-night variability. This pattern overlaps with the changes seen in resting heart rate and HRV, making multi-metric trend reading more informative than any single number.

---

What a persistently elevated overnight rate suggests

There is meaningful clinical data connecting nocturnal respiratory rate to longer-term health outcomes — and it is worth understanding both what that evidence says and what it does not.

A large longitudinal study by Baumert and colleagues — combining 2,686 men from the MrOS Sleep Study and 406 women from the Study of Osteoporotic Fractures — used overnight polysomnography to measure mean nocturnal respiratory frequency. After multivariable adjustment, a nocturnal respiratory rate at or above 16 breaths per minute was independently associated with greater cardiovascular and all-cause mortality risk in both cohorts.

Several important caveats apply before interpreting this in a wearable context. The study population was older (the MrOS cohort was aged 65+), the measurements were made via clinical polysomnography rather than wrist PPG, and the threshold of 16 breaths per minute was derived from that specific population's distribution — it is not a universal danger line. What the finding does support is the directional relationship: chronically elevated nocturnal respiratory rate, as part of a broader physiological pattern, is a signal worth attending to. It tells you something about the load your cardiorespiratory system is carrying overnight — not what you have or don't have.

For most people using a wearable, the operationally relevant signal is not whether you're at 14 or 16, but whether your rate has shifted persistently from your own baseline — especially if it's moving in the same direction as other metrics like resting heart rate and HRV.

---

The limits: what wrist PPG can and cannot detect

The RSA-derived respiratory rate that Apple Watch measures is a useful directional indicator. It is not a clinical measurement, and its error margin matters at the level of individual breaths.

The Natarajan et al. validation showed a mean absolute error of 0.46 breaths per minute against polysomnography — which is better than typical manual clinical counting (where inter-rater variability can exceed 2 breaths per minute). But the measurement degrades when the PPG signal is noisy, the watch fits loosely, or sleep is fragmented. For trend tracking over weeks — which is the appropriate use case — these limitations are manageable; the noise averages out.

Regarding sleep apnea specifically: Apple Watch does include a sleep apnea notification feature, introduced with watchOS 11. This uses accelerometer data to detect breathing disturbances consistent with moderate-to-severe obstructive sleep apnea, and it is a distinct system from the general respiratory rate tracking. Consumer wrist PPG-based respiratory rate is not designed to screen for sleep-disordered breathing. If you have symptoms of sleep apnea — snoring, gasping, daytime sleepiness, morning headaches — a clinical evaluation remains the appropriate path, not a wearable reading. Apple's own materials state explicitly that Vitals app measurements are not intended for medical use.

---

Where Sam Health fits in

Sam tracks your overnight respiratory rate alongside your other vitals and establishes your personal baseline — the pattern that holds steady when you're sleeping well and living without unusual stressors. When your respiratory rate rises above that baseline for several consecutive nights, Sam surfaces the pattern and the context: was there alcohol involved? A training spike? An elevation change? Or nothing obvious at all?

The goal is not to generate alerts from a single data point. It is to give you a consistent, personalised view of a metric that genuinely shifts before you notice many things have changed — so that when it does, you see it in context rather than discovering it in retrospect. For a complete overview of the wearable metrics Sam works with, see the wearable biomarkers that actually matter.

Try Sam Health

Sources

1. Apple Inc. Track your overnight vitals with Apple Watch. Apple Support. support.apple.com/en-us/120142 (accessed 16 May 2026).

1. Apple Inc. Sleep apnea notifications on your Apple Watch. Apple Support. support.apple.com/en-us/120031 (accessed 16 May 2026).

1. Natarajan A, Su HW, Heneghan C, et al. Measurement of respiratory rate using wearable devices and applications to COVID-19 detection. NPJ Digit Med. 2021;4(1):136. doi:10.1038/s41746-021-00493-6 (accessed 16 May 2026).

1. Douglas NJ, White DP, Pickett CK, Weil JV, Zwillich CW. Respiration during sleep in normal man. Thorax. 1982;37(11):840–4. PMID: 7164002 (accessed 16 May 2026).

1. Gutierrez G, Williams J, Alrehaili G, et al. Respiratory rate variability in sleeping adults without obstructive sleep apnea. Physiol Rep. 2016. PMID: 27597768 (accessed 16 May 2026).

1. Baumert M, et al. Mean nocturnal respiratory rate predicts cardiovascular and all-cause mortality in community-dwelling older men and women. Eur Respir J. 2019;54(1):1802175. PMC: 7864583 (accessed 16 May 2026).

1. Kolla BP, Foroughi M, Saeidifard F, et al. The impact of alcohol on breathing parameters during sleep: a systematic review and meta-analysis. Sleep Med Rev. 2018;42:59–67. doi:10.1016/j.smrv.2018.05.007 (accessed 16 May 2026).