Risk of hearing damage from infant white noise: a reading of the evidence

What we know: device output measurements

The most-cited piece of evidence in this space is the 2014 audiology study by Hugh, Cutler, and El-Naga, published in Pediatrics. The team measured the dB output of 14 commercially available infant sound machines at three distances (30 cm, 100 cm, 200 cm) across all available volume settings. All 14 machines produced more than 50 dB at 30 cm at maximum volume; three produced more than 85 dB at maximum at the closest distance. 85 dB is the level at which NIOSH establishes adult occupational hearing protection requirements for eight-hour exposures. Infant cochleas are more vulnerable than adult ones and sleep windows are longer than eight hours per night. The Hugh study did not measure hearing outcomes; it measured device capability. Its policy importance lies in showing that consumer hardware can plausibly produce risky output under realistic but extreme use patterns. See the dedicated Hugh 2014 explainer for the full methodology.

Subsequent replication and extension of the Hugh measurements has confirmed the broad pattern. A 2020 review in the Journal of Clinical Sleep Medicine compiled measurements across newer machine generations and found that while average output had dropped (manufacturers responded to Hugh), the spread of available output remained wide enough that maximum-volume settings on multiple devices still exceeded 80 dB at typical nursery placement distances. Newer products, including those marketed as “app-controlled” and “baby-safe,” can still exceed safe levels if a user disables the safety cap.

On the threshold side, NIOSH's recommended exposure limit for adults is 85 dB time-weighted over eight hours, with the limit halving (in permissible exposure time) for every additional 3 dB. The CDC's noise-induced hearing loss reference notes that prolonged exposure above 70 dB carries risk for adults, and that pediatric thresholds should be lower. The American Academy of Pediatrics 2023 policy statement (see AAP 2023 summary) accordingly recommends 50 dB as the infant sound machine ceiling, a substantial safety margin below any documented adult harm threshold.

What we do not know: long-term cohort data

There is no published prospective cohort study following infants exposed to white noise machines over months or years and measuring later-life hearing outcomes. This is a notable evidence gap. The reason it persists is partly practical (the necessary follow-up is decades long and confounded by other noise exposures) and partly ethical (you cannot randomise infants to risky use patterns to test the hypothesis). The available evidence is therefore a chain of inference: measured device output, extrapolation from adult occupational data, and clinical acoustics of the developing cochlea.

That gap matters in two directions. First, it means parents who follow the AAP guidance can be reasonably confident in a wide safety margin without claiming certainty of zero risk. Second, it means alarmist coverage that asserts white noise machines “cause” hearing damage is overstating what the evidence supports. Hugh 2014 demonstrates capability of harm at extreme settings, not realised harm at normal use.

The closest indirect evidence comes from NICU acoustic research, which has documented worse developmental outcomes in higher-noise NICU environments. The NICU pattern is not directly translatable to the home setting, but it informs the precautionary direction the AAP statement takes.

Sound machine versus other ambient noise

One useful comparison: the noise environment a baby experiences at home without a sound machine. Urban apartments often sit at 45 to 55 dB ambient during waking hours and 35 to 45 dB overnight. Traffic, neighbours, ventilation, refrigerators, and conversation all contribute. A baby is therefore not in a noise vacuum even without a machine. The relevant question is not white noise versus silence; it is white noise at a chosen level versus whatever the ambient is.

In many real settings, a sound machine at 45 dB (just below the AAP ceiling) is comparable to or slightly above the ambient. The masking effect is what makes a difference to sleep onset, not a step change in total exposure. The AAP recommendation is conservative because it does not assume the ambient is silent; it caps the total at a level below both adult occupational concern and the typical urban nighttime ambient with a safety margin.

By contrast, vacuum cleaners (around 70 dB), hairdryers (75 to 85 dB), and many household appliances exceed the AAP infant ceiling at typical proximity. Brief exposures at these levels are not equivalent in risk to all-night exposure at the same level: noise risk integrates dose over time. The AAP's focus on sound machines reflects the unique combination of moderate level + long duration + close proximity that characterises overnight nursery use.

The precautionary principle and what it means practically

The precautionary principle in clinical settings is, in plain terms, when the cost of being cautious is low and the cost of being wrong is high, default to caution. For infant sound machine use, the cost of being cautious (a machine 7 ft from the crib at 45 dB instead of 6 inches at 65 dB) is essentially zero: the baby still gets masking, the parents still sleep. The cost of being wrong (an unmeasured cumulative exposure over months that turns out to nudge later hearing thresholds) is high and irreversible. The precautionary calculus therefore favours conservative settings.

In practice this means three things. First, measure with a phone app rather than trusting the volume knob. Second, place the machine at the back of the nursery, not in the crib. Third, use a sleep timer where the baby tolerates it (see all-night vs sleep timer) rather than continuous overnight playback by default.

The other side of the precautionary principle: it is also cautious not to assume the worst about a tool that helps families sleep. Sleep deprivation in new parents carries documented health and safety risks. A sound machine used within AAP guidance is a net positive intervention, not a hidden harm. The risk framing is about edge cases (maximum volume, crib placement, all-night every night for years), not about the practice itself.

If you are worried, what to actually do

Three concrete actions cover the realistic worry space.

One, measure your setup once with the NIOSH SLM app (iOS) or Decibel X (Android). Place the phone at the baby's head position and read the LAeq over 30 to 60 seconds. If you are at or below 50 dB, you are within AAP guidance. If you are above, lower the volume or move the machine further away.

Two, attend the standard well-child hearing screens. Newborn hearing screening is universal in U.S. births. Subsequent audiological evaluation is recommended at 9, 18, and 30 months if there are concerns, and at age 4 for routine screening, per AAP Bright Futures. If sound machine use has been within guidance, additional dedicated screening is not indicated.

Three, if there is any clinical reason to suspect a hearing issue (delayed speech, lack of startle response to loud sounds, family history), raise it with your pediatrician. They will refer to a pediatric audiologist for formal age-appropriate testing. Sound machine use, even at high volume, is rarely the leading hypothesis for clinically detected loss; genetic, perinatal, and infectious causes dominate.

Frequently asked

Sources

• American Academy of Pediatrics, Noise: A Hazard for the Fetus and Newborn (Update), Pediatrics 2023
• Hugh SC, Cutler N, El-Naga A et al. Pediatrics 2014; 133(4): 677-681
• NIOSH occupational noise reference, cdc.gov/niosh
• CDC noise-induced hearing loss page, cdc.gov
• NIDCD “Noise-Induced Hearing Loss” reference, nidcd.nih.gov