Does white noise help you sleep? What 38 studies found
The review
In 2021, sleep researchers at the University of Pennsylvania published a systematic review in Sleep Medicine Reviews of every study they could find on continuous white noise, or similar broadband noise, and sleep. They found 38 studies. Some were in sleep labs, some in hospitals, some at home; some measured sleep with an EEG, others asked people how they slept.
Their conclusion was blunt. The studies varied in the kind of noise, its level, how sleep was measured and what it was compared with, and the results varied with them, “with the extremes being that continuous noise improves or disrupts sleep”. Rated by the standard GRADE criteria, the quality of evidence that continuous noise improves sleep was very low, which the authors noted contradicts how widely it is used. They called for better studies before noise is promoted as a sleep aid, “especially since it may also negatively affect sleep and hearing”.
Where noise seems to help
The most convincing studies are about covering other sounds.
- A noisy ward, 2005. In a small sleep-lab study in Sleep Medicine, 4 healthy people slept with recorded intensive-care noise, then with the same noise plus white noise. The ward noise more than tripled their arousals from sleep; adding white noise brought them back to near normal. What woke them was the jump from background to peak, about 17 to 18 dB, not the loudness of the peak. Steady noise raised the background so the jumps were smaller.
- Falling asleep early, 2017. In a randomised trial in Frontiers in Neurology, 18 healthy adults were sent to bed 90 minutes early, a model of a night when sleep will not come. With broadband sound at about 46 dB from two speakers, they reached stage 2 sleep sooner than in the room’s normal background of about 40 dB: a median reduction of 38%.
Both are small, short and in healthy people. They support noise as a mask for a noisy room, and perhaps as a cue that it is time to sleep, more than as something that improves sleep by itself. How masking works is on its own page.
Where noise may hurt
- Sleep structure. In a 2026 sleep-lab study of 25 adults in the journal Sleep, steady pink noise at 50 dBA all night cut REM sleep by about 19 minutes compared with a quiet night, and adding it to traffic and aircraft noise worsened sleep structure overall. Earplugs did better. See steady noise and REM sleep.
- Hearing. Noise played loudly, close to the ear, for eight hours a night adds up. A 2014 study in Pediatrics found every one of 14 infant sleep machines could exceed 50 dBA at 30 cm, the nursery limit, and three passed 85 dBA.
What this means at night
- Use noise for a reason. A noisy street, a snoring partner, a hotel corridor. In a quiet room there is little evidence it adds anything.
- Keep it quiet. Just loud enough to soften the sounds you want to cover. How loud should white noise be? gives practical levels.
- Let it fade. If it helps you fall asleep, you may not need it at 4 a.m. A timer that fades it once you are asleep is a fair compromise; see should white noise play all night?
- Pick a colour you like. White noise is the hissiest. Pink and brown are deeper and many people find them easier to sleep to. No colour has better evidence.
- Not near a baby’s ears. Keep any noise far from a cot and quiet.
Press play above for white noise made live in your browser: it never loops, and you can set a timer that fades it out.
Where Nightjar helps, and where it doesn’t
Nightjar’s noise beds are quiet, fade out with the timer, and can follow the room rather than run loud all night. With Adapts to your room, the noise bed rises only when the room gets louder, then settles; the microphone is read for level only and nothing is recorded, stored or sent. The sleep timer runs from 5 minutes to 10 hours, or Likely asleep, or Keep playing, and the sound fades out gently. Its own note on noise beds is labelled Mixed and says what this page says: noise can cover other sounds, and the evidence that it improves sleep itself is weak.
Nightjar does not claim noise improves your sleep. It is a tool for a noisy room, used quietly; in a quiet room, the honest advice is that you may not need it.
Sources
- Riedy, Smith, Rocha and Basner, Noise as a sleep aid: a systematic review, Sleep Medicine Reviews, 2021
- Stanchina and colleagues, The influence of white noise on sleep in subjects exposed to ICU noise, Sleep Medicine, 2005
- Messineo and colleagues, Broadband sound administration improves sleep onset latency in healthy subjects in a model of transient insomnia, Frontiers in Neurology, 2017
- Basner and colleagues, Efficacy of pink noise and earplugs for mitigating the effects of intermittent environmental noise exposure on sleep, Sleep, 2026
- Hugh and colleagues, Infant sleep machines and hazardous sound pressure levels, Pediatrics, 2014
Last checked 13 November 2026.
Questions
Is it bad to sleep with white noise every night?
There is no clear evidence of harm at a low level, but no clear evidence of benefit either, and a 2026 lab study found steady noise reduced REM sleep. Keeping it quiet and letting it fade after you fall asleep limits any downside.
Is white, pink or brown noise better for sleep?
No colour has been shown to be better. Many people find white noise hissy and prefer the deeper sound of pink or brown, but that is a matter of comfort, not evidence.
Is white noise safe for babies?
Use great care. A 2014 study found all 14 infant sleep machines it tested could exceed the recommended noise limit for hospital nurseries at 30 cm. Keep any machine far from the cot, at a low volume, and ask your health visitor or doctor.