Noise and Sleep: What WHO Guidance Actually Measures

WHO's environmental noise guidance for the night period uses Lnight,outside: an A-weighted equivalent level over the night, averaged across a long period such as a year, measured outside the dwelling.

Every part of that definition matters. Outside means it does not describe your bedroom. Long-term average means it does not describe last night. Population-level means it is designed to inform policy on roads, railways and airports, not to tell an individual what their room should measure.

Turning an outdoor level into an indoor one requires knowing the sound insulation of the specific façade, whether windows are open, and where the source is. Those vary by tens of decibels between buildings.

So a page that presents a WHO night value as "the recommended bedroom level" has performed a conversion that no source supports. We do not do it.

The mechanisms are reasonably well described: noise can cause awakenings, brief cortical arousals that you do not remember, autonomic responses such as heart-rate changes, and shifts between sleep stages.

Character matters as much as level. An intermittent sound with sharp onsets — a door, a vehicle passing, a snore — disturbs sleep differently from a steady sound at the same average level. This is one reason a single decibel limit does a poor job of describing sleep disturbance.

Because the useful variable is often variability rather than absolute level, the interventions that help are the ones that reduce sharp events: sealing the gap under a door, moving the bed away from a shared wall, closing a window on the road side, or masking intermittent noise with something steady.

Measuring is genuinely useful here — not for comparison against a guideline, but for comparing your own options. Window open against closed, fan on against off, bed position A against B. Those are relative comparisons, which is what an uncalibrated measurement is actually good for.

We could not verify, from the primary guideline documents, exact numeric night-noise values in a form we are willing to quote, and we found no authoritative indoor bedroom decibel limit at all.

We also found no published evidence for the widely repeated claim that a specific bedroom level guarantees undisturbed sleep. Since the disturbance depends on sound character and individual sensitivity, we would be surprised if such a figure could exist.

There is a second gap worth naming. Most sleep-noise research studies transport noise — roads, railways, aircraft — because that is what policy needs. Domestic sources such as a partner snoring, a heat pump cycling or a neighbour's television are far less studied, so the sounds people actually complain about at home are the ones with the thinnest evidence base.

Where that leaves an individual is with comparison rather than compliance: measure your own options against each other, and treat any absolute figure as an estimate rather than a verdict.