New Guidance on Noise and Sleep Quality, and What It Means for You

In February 2026, a University of Pennsylvania sleep-lab trial published in SLEEP found that foam earplugs nearly eliminated the deep-sleep loss caused by recorded traffic and aircraft noise, while pink noise played at 50 dBA cut REM sleep. The official limits have barely moved: HUD's June 2026 rule revision kept its noise thresholds, the FAA's review of its 65 dB aircraft noise threshold is on hold until an advisory committee reports, and the WHO's night limits still date from 2018. For most readers, the practical change is in the bedroom rather than the rulebook: earplugs now have the stronger new evidence, while the evidence on sound machines is mixed.

Key takeaways

• The WHO's 2018 Environmental Noise Guidelines for the European Region are still the current health-based values: outdoor night averages below 45 dB for road traffic, 44 dB for rail and 40 dB for aircraft. The WHO made no night recommendation for wind turbines because the evidence was too weak.
• The FAA still uses a yearly average of 65 dB DNL as its threshold for significant aircraft noise. Its noise policy review is waiting on an advisory committee whose members were still being recruited in summer 2026.
• HUD's noise rule (24 CFR part 51, subpart B) was revised on June 12, 2026, and corrected on August 20, 2026. The changes are administrative. Sites above 65 dB DNL are still "normally unacceptable" for HUD-assisted housing, and sites above 75 dB are "unacceptable".
• In a Penn-led trial of 25 adults published in SLEEP in 2026, foam earplugs nearly eliminated the loss of deep sleep caused by recorded traffic and aircraft noise. Pink noise at 50 dBA reduced REM sleep by about 19 minutes.
• Yearly averages such as Lnight hide the loud single events that wake people. A phone app can log those events but cannot give an official Lnight.

What changed and what it means

What it means for you

• Near an airport, highway or rail line: no official limit has changed. The FAA still uses 65 dB DNL and has given no date for a decision on its review, and the WHO's outdoor night values (below 40 dB for aircraft, 45 dB for road traffic and 44 dB for rail) are the 2018 figures. A one-week log, plus your airport's noise contour map if it publishes one, tells you more about your own nights. See "What FAA and HUD rules cover" and the checklist at the end.
• Renting: HUD's 2026 revision changed who approves projects, not the thresholds, and its noise rule is not a tenant's right. Noisy neighbors and late-night construction fall under your lease and state and local rules. Removable window inserts and an app log of event times are covered in "Bedroom fixes, including renters".
• Using earplugs or a sound machine: in the 2026 Penn trial, earplugs nearly eliminated the effects of noise but protected less at the loudest peaks, and how much they blocked varied widely between people. Pink noise at 50 dBA cut REM sleep, though a smaller pilot study was more favorable. Details are in "Pink noise and earplugs: new trials".
• Parents of babies and young children: the editorial accompanying the trial urged caution about broadband noise for children, and a 2014 study found that infant sleep machines at maximum volume can exceed hospital nursery limits. See "Are white noise machines safe for babies?" in the FAQ below.
• Sleeping badly most nights: earplugs and quieter windows won't fix sleep apnea, and poor sleep that affects your days, or trouble sleeping three or more nights a week that goes on for months, is a reason to see a clinician. See "When to talk to a clinician".

Who issued what, and when

Each item below was checked against the issuing body's own page or the journal article on October 2, 2026.

Several bodies have issued nothing new. EPA closed its Office of Noise Abatement and Control after the federal government concluded in 1981 that noise was best handled by state and local governments. Its noise page still points to health materials from 1981 and 1991. We found no noise-specific sleep guidance from CDC or NIH in 2024–2026. The American Academy of Sleep Medicine's list of position statements includes none on environmental noise.

How WHO set its night limits

The WHO Environmental Noise Guidelines for the European Region were launched on October 10, 2018. They are written for policymakers, not as limits for individual homes. The values are outdoor yearly averages at the most exposed wall of a home, and the road, rail and aircraft night values are rated "strong" recommendations.

The WHO picked each night value at roughly the level where no more than about 3% of residents report being highly sleep-disturbed. It chose 3% to match its earlier 2009 Night Noise Guidelines, which set a 40 dB outdoor night target and a 55 dB interim target. The guidelines' own tables show that the same average disturbs people very differently depending on the source:

Even at 40 dB, the aircraft figure is well above 3%. The review's models start at 40 dB because noise predictions below that level are less accurate. So the aircraft value marks the bottom of the evidence, not a level shown to be safe. A 2022 update of the WHO sleep review added 11 studies with more than 109,000 responses. It matched the WHO's estimates at 40 dB, found more disturbance from aircraft at high levels, and concluded the night values did not need revisiting.

Europe is far from meeting these values. The EEA's 2025 noise report estimates that more than 30% of Europeans are exposed above WHO levels. It also finds that the number of people highly annoyed by transport noise fell only about 3% from 2017 to 2022, against an EU goal of a 30% cut by 2030.

Lnight, Lden, DNL and LAmax

All of these use A-weighted decibels (dBA), which weight frequencies roughly the way human hearing does. The scale is logarithmic: two equal sources together are about 3 dB louder than one.

• Lnight is the average sound level over the eight night hours, usually 11 p.m. to 7 a.m., averaged across a year.
• Lden is a 24-hour European average with a 5 dB penalty for evening hours and 10 dB for night hours.
• DNL (day-night average sound level) is the US version used by the FAA and HUD. Per HUD's noise regulation, it adds 10 dB between 10 p.m. and 7 a.m., with no evening penalty.
• LAmax is the loudest moment of a single event, such as a truck, train or overflight.

Lden = 10 × log10( [12 × 10^(Ld/10) + 4 × 10^((Le+5)/10) + 8 × 10^((Ln+10)/10)] / 24 )

Ld = average level, day (12 hours, typically 7 a.m.–7 p.m.)
Le = average level, evening (4 hours, typically 7–11 p.m.)
Ln = average level, night (8 hours, typically 11 p.m.–7 a.m.)

Illustrative example: a street averages 55 dB by day, 52 dB in the evening and 50 dB at night. A plain energy average of those three periods is 53.4 dB. With the penalties, Lden comes out at 57.6 dB. That is 4.2 dB higher. Most of the rise comes from the night penalty: applied on its own, it adds about 3.8 dB.

Why averages can mislead

Averages measure total sound energy, not how often a sleeper is jolted. The WHO says LAmax-type measures suit night rail and aircraft noise, because peaks drive awakenings, but it set no single-event limit because the long-term evidence was too uncertain.

Illustrative example: take an eight-hour night with a 40 dB outdoor background. In Night A, 40 vehicles each reach 75 dB for 10 seconds. In Night B, only four do, but each reaches 85 dB. Both nights average 56.5 dB. A steady 56.5 dB hum all night would give the same figure. The events fill 1.4% of Night A and 0.14% of Night B, yet they set the average. Doubling Night A to 80 events raises it only 3 dB, to 59.5 dB. An "Lnight of 55" on a map can therefore describe very different nights.

What a phone decibel app can show

An app on a nightstand measures indoor sound over a few hours. Lnight, Lden and DNL figures on noise maps are outdoor yearly averages at the facade, so the two numbers aren't directly comparable.

Accuracy is the second limit. CDC's NIOSH Sound Level Meter app is rated accurate within ±2 dBA and reports LAeq averages and maximum levels. It is iOS-only, because NIOSH says it can't verify Android apps across so many different devices. NIOSH's lab validation, which found readings within ±1 dBA, used calibrated external microphones at 65–95 dBA, a range that covers workplace noise. That figure applies to calibrated external microphones, not a phone's built-in one, and a quiet bedroom at 30–40 dB is below that range, so low readings from a bare phone are less certain.

What FAA and HUD rules cover

FAA: 65 dB DNL, still under review

The FAA treats a yearly average of 65 dB DNL as the threshold for significant aircraft noise in environmental reviews and as the limit for compatible residential land use. Its Noise Policy Review is examining whether to keep that threshold and whether to use other metrics alongside DNL. The review drew 4,857 public comments between May 1 and September 29, 2023. Because of the 2024 reauthorization law, the FAA has put off any decision until the Aircraft Noise Advisory Committee issues its report. The FAA's July 10, 2026 notice asked for committee nominations. The FAA has given no date for a decision.

The review grew partly out of the FAA's Neighborhood Environmental Survey, published in January 2021. Among more than 10,000 residents near 20 airports, more people were highly annoyed at every noise level studied than the 1970s-era Schultz curve predicted. The survey measured annoyance, not sleep, although most phone follow-up respondents who were likely to be annoyed said aircraft had woken them at home.

HUD: thresholds unchanged after the 2026 revision

HUD's June 12, 2026 direct final rule took effect July 13, 2026. In "unacceptable" noise zones (above 75 dB DNL), it lets the HUD program office funding a project approve the project, its noise attenuation measures and any waiver of the environmental impact statement. Before, only the Office of Community Planning and Development could do this. The rule also drops noise surveillance and contour-review provisions that HUD says it never put into practice. An August 20, 2026 correction fixed a drafting error. The standards themselves did not change:

• Acceptable: 65 dB DNL or below. This can be raised to 70 dB in special circumstances.
• Normally unacceptable: above 65 dB and up to 75 dB. Projects need a Special Environmental Clearance (a full environmental impact statement in some cases, such as largely undeveloped areas) and at least 5 dB of extra sound attenuation (10 dB above 70 dB).
• Unacceptable: above 75 dB. Projects require an environmental impact statement, which can be waived in limited cases.
• Goals: HUD aims for 55 dB DNL outdoors, the level EPA recommends for residential areas, and 45 dB DNL indoors, with emphasis on bedrooms.

The rule covers HUD programs that assist, subsidize or insure housing, manufactured home parks, nursing homes, hospitals and land development for housing. It is not a tenant's right, and it doesn't apply to private rentals without HUD involvement. Noise alone won't stop HUD support for buying or reselling an otherwise acceptable existing building, but HUD treats noise as a marketability factor when it decides how much insurance or other assistance to give.

Pink noise and earplugs: new trials

The largest new study is a seven-night lab trial led by Mathias Basner at the University of Pennsylvania, published in SLEEP. Twenty-five healthy adults aged 21 to 41 slept under brain-wave monitoring while researchers played recordings of jets, helicopters, drones, a low sonic boom, a train and a car, adjusted to sound as they would inside a bedroom, with peaks of 45–65 dBA. Some nights added pink noise (a steady hiss with more low-frequency energy than white noise), and some added foam earplugs.

• Environmental noise cut deep (N3) sleep by about 23 minutes.
• Pink noise alone at 50 dBA cut REM sleep by about 19 minutes.
• Adding pink noise to environmental noise made sleep structure worse overall, despite small gains in some fragmentation measures.
• Earplugs nearly eliminated the effects of environmental noise, but protected less at 65 dBA peaks.
• Morning cognition, heart measures and hearing didn't differ; participants rated sleep worse after noise and pink noise nights.

An accompanying editorial noted that earplugs blocked very different amounts of sound for different people: about 25 dB on average, but anywhere from about 11 to 42 dB. It urged caution about widely promoting broadband noise, particularly for vulnerable groups such as children, because REM sleep matters for brain development.

A smaller pilot study in Communications Medicine reached a more favorable view of pink noise. In 12 healthy adults, continuous pink noise at 45 dB reduced the sleep fragmentation and blood-chemistry changes caused by traffic noise. Its authors urged caution given the small, similar sample. Both studies saw pink noise ease some noise-related fragmentation. They differ on the net effect, and in size, pink-noise level and outcomes measured, so neither settles the question. They line up with a 2021 systematic review of 38 studies, which rated the evidence that continuous noise improves sleep as very low quality.

Some official advice predates this work. A CDC training module for shift-working nurses, last reviewed in 2020, suggests earplugs or a fan or "white noise" machine to mask noise. It hasn't been updated to reflect the 2026 trial.

Bedroom fixes, including renters

A window is the weakest part of most bedroom walls. The WHO guidelines estimate that indoor levels run about 10 dB below outdoor levels with the window open, 15 dB with it tilted or half open, and about 25 dB with it closed. On a street with a 55 dB outdoor night average, that works out to roughly 45, 40 and 30 dB indoors.

1. Try earplugs properly. NIOSH's roll-pull-hold method for foam plugs is to roll the plug thin, pull the top of the ear up and back, and hold the plug while it expands. Your voice should sound muffled. Earplugs also muffle alarms, so check that you can still hear your smoke alarm with them in.
2. Close and seal the window if you can stay cool. Sound leaks through any air gap, so weatherstripping helps.
3. Sleep on the quiet side. The WHO review found that homes with a quiet side had less reported trouble falling asleep, although it rated that evidence very low quality.
4. Be cautious with sound machines. The 2026 evidence on pink noise is mixed. The REM loss in the larger trial came at 50 dBA, and no level has been shown to be safe for long-term use. The editorial urged particular caution for children.
5. Renters: pressure-fit acoustic window inserts and heavy door sweeps can be removed when you move out, but check your lease before installing anything. State and local noise rules and your lease, not federal rules, cover noisy neighbors and late-night construction. A log of event times from an app supports a complaint.

When to talk to a clinician

The National Heart, Lung, and Blood Institute defines chronic insomnia as trouble sleeping three or more nights a week for more than three months that another health problem doesn't fully explain. It advises seeing a provider when poor sleep affects daily life, with a 1–2 week sleep diary. Loud snoring, gasping or pauses in breathing are common signs of sleep apnea. Earplugs and quieter windows won't fix sleep apnea. This article is general education, not medical advice. A clinician can say whether noise or something else is the main problem.

Frequently asked questions

Can you get used to sleeping with noise?

Partly. A 2014 review in The Lancet led by Mathias Basner found that people tend to get used to noise, but how much differs from person to person and is rarely complete. Reactions such as a jump in heart rate fade much less with repeated exposure than brain-wave arousals do. The authors see this as a likely reason night noise might matter more for long-term heart health than daytime noise.

Can noise affect your sleep without waking you up?

Yes. The same Lancet review reports that single noise events peaking as low as 33 dB LAmax can set off physiological reactions during sleep, including a faster heartbeat, body movements and brain-wave arousals. Full awakenings are only one of the responses it lists.

Is night-time noise linked to heart problems?

Studies link long-term traffic noise to heart disease, although they show associations across populations, not a proven cause in any one person. Meta-analyses summarized in the Lancet review found that the risk of high blood pressure and ischemic heart disease, including heart attack, rose by 7% to 17% for every 10 dB increase in average road traffic or aircraft noise. Those estimates were adjusted for age, sex, smoking, body-mass index and other factors. The review adds that night exposure might matter more than daytime exposure.

How quiet should a bedroom be for sleep?

The WHO's 1999 Guidelines for Community Noise set indoor bedroom values of 30 dB LAeq for continuous noise and 45 dB LAmax for single events. They recommend a lower value for noise with a lot of low-frequency sound and an even lower single-event limit for sensitive people. WHO's 2018 guidelines for Europe supersede the 1999 edition there but recommend that its indoor values remain valid. Because these are indoor figures, they are closer to what a bedside app records than the outdoor yearly averages above, though phone readings near 30 dB are less certain.

Are white noise machines safe for babies?

A 2014 study in Pediatrics tested 14 infant sleep machines at maximum volume. At 30 cm (about a foot), all of them exceeded 50 dBA, the current recommended noise limit for infants in hospital nurseries. Three produced more than 85 dBA, a level that, played for more than eight hours, would exceed adult workplace limits for accumulated noise and risk noise-induced hearing loss. The authors concluded the machines can produce levels that may damage infant hearing. If you use one, ask your pediatrician about a safe volume and placement.

A one-week noise checklist

• If a nearby airport publishes noise contour maps, check whether your home falls inside its DNL 65 line. Also note any busy road or rail line near your bedroom.
• For two or three nights, run a sound-level app in the same spot, once with the window closed, and note event counts, peaks and times.
• Try fitted foam earplugs for several nights, and check that you can still hear your smoke alarm.
• Renters: put repeated noise problems in writing to the landlord with your log, and check your city's noise ordinance.
• If poor sleep continues three or more nights a week, or if someone hears you gasping or pausing in your breathing, book a visit and bring your sleep diary.

Sources

• WHO Regional Office for Europe: Environmental Noise Guidelines for the European Region (2018)
• WHO Europe: New WHO noise guidelines for Europe released (October 10, 2018)
• WHO Regional Office for Europe: Night Noise Guidelines for Europe (2009)
• WHO: Guidelines for Community Noise (Berglund, Lindvall and Schwela, 1999)
• Basner et al.: Auditory and non-auditory effects of noise on health (The Lancet, 2014)
• Smith, Cordoza and Basner: Environmental noise and effects on sleep, an update to the WHO systematic review (Environmental Health Perspectives, 2022)
• WHO Regional Office for Europe: Disability weights for noise-related health states in the WHO European Region (May 27, 2024)
• European Environment Agency: Environmental noise in Europe 2025 (EEA Report 05/2025)
• FAA: Noise Policy Review
• Federal Register: Aircraft Noise Advisory Committee, 91 FR 42788 (July 10, 2026)
• FAA: Neighborhood Environmental Survey
• eCFR: 24 CFR part 51, subpart B, Noise Abatement and Control
• Federal Register: Revising HUD's Noise Abatement and Control Regulations, 91 FR 35624 (June 12, 2026)
• Federal Register: Correcting amendment, 91 FR 53717 (August 20, 2026)
• EPA: Clean Air Act Title IV, Noise Pollution
• American Academy of Sleep Medicine: Position statements
• Basner et al.: Efficacy of pink noise and earplugs for mitigating the effects of intermittent environmental noise exposure on sleep (SLEEP, 2026)
• Garcia Molina: Protecting sleep in a noisy world, editorial (SLEEP, 2026)
• Vincens et al.: Pink noise reduces impact of traffic noise on sleep and the blood metabolome, a cross-over pilot study (Communications Medicine, 2026)
• Riedy et al.: Noise as a sleep aid, a systematic review (Sleep Medicine Reviews, 2021)
• CDC/NIOSH: Training for nurses on shift work, create a good sleep environment
• CDC/NIOSH: Sound Level Meter app
• Sun et al.: The potential use of a NIOSH sound level meter smart device application in mining operations (Noise Control Engineering Journal, 2019)
• CDC/NIOSH: How to wear soft foam earplugs
• NHLBI: Insomnia
• NHLBI: Insomnia diagnosis
• NHLBI: Sleep apnea symptoms
• Hugh et al.: Infant sleep machines and hazardous sound pressure levels (Pediatrics, 2014)