What the Noise Does

The World Health Organization estimates that between one and 1.6 million disability-adjusted life years are lost in Western Europe every year due to road traffic noise. That is a larger burden than lead exposure. It is comparable to the toll attributed to fine particulate matter in countries where air quality monitoring has driven decades of policy reform. And yet noise, as a chronic environmental health exposure, remains largely absent from the clinical encounter. It is not routinely assessed, not routinely documented, and not incorporated into cardiovascular or cognitive risk profiles in the way that blood pressure, lipid levels, or sleep apnoea are.

That absence is not a reflection of the evidence. It is a reflection of how medicine organises its attention.

What the Evidence Now Shows

A 2025 longitudinal analysis published in Behavioral Sciences, drawing on three waves of the China Health and Retirement Longitudinal Study (CHARLS) and 3,459 individuals aged 45 and above, found a significant inverse relationship between noise pollution exposure and overall cognitive function. Each unit increase in the noise pollution index corresponded to a 0.41-point reduction in cognitive scores. The study identified two principal mediating mechanisms: disrupted sleep and depressive symptoms. Neither is surprising. Both are well-documented downstream effects of chronic noise exposure. What the CHARLS analysis adds is longitudinal rigour across a large middle-aged and older population — the cohort in which cognitive trajectory is most consequential.

The mechanism is not primarily hearing damage. Noise-induced hearing loss is a separate, parallel consequence of chronic high-decibel exposure. The cognitive and cardiovascular pathways implicated in ambient noise research operate at levels well below those required to damage the auditory apparatus. The mechanism involves the autonomic nervous system: chronic noise activates the hypothalamic-pituitary-adrenal axis, elevates cortisol and adrenaline, and produces a sustained low-grade stress response. This pathway is the same one through which psychological stress — work pressure, financial insecurity, relationship difficulty — produces cardiovascular disease. Noise is, in this sense, a form of involuntary stress.

Noise is not merely unpleasant. It is a chronic stressor delivered without consent, distributed by geography, and concentrated in the bodies of those who cannot afford to live elsewhere.

The cardiovascular evidence is more established than the cognitive evidence, and it is already alarming. A systematic review and meta-analysis published in the European Heart Journal found that long-term exposure to road traffic noise was associated with an increased risk of ischemic heart disease, stroke, and heart failure. The dose-response relationship held at exposure levels common across major urban areas: 50 to 65 decibels, which describes the ambient sound environment of an ordinary city street, not a construction site or a motorway.

The Geography of Exposure

Noise exposure is not evenly distributed. Urban noise — traffic, aviation, rail, industrial activity — is concentrated in specific zones, and those zones correspond closely with the geography of economic disadvantage. In European cities, lower-income neighbourhoods are disproportionately located near major roads and flight paths. In American cities, the correlation between highway proximity and historical redlining has been documented across multiple metropolitan analyses. The consequence is that chronic noise exposure, like chronic air pollution, operates as an environmental amplifier of existing health inequality. Those least able to absorb its consequences bear the largest doses.

This is the dimension of the noise evidence that makes the clinical framing — treat it as individual sensitivity, manage with earplugs, recommend white noise machines — structurally inadequate. The individual clinical response is not wrong. It is simply beside the point for a burden that is overwhelmingly environmental in origin and social in distribution.

What Has Dementia Research Added

A systematic review published in medrxiv.org in 2025, drawing on MEDLINE, EMBASE, CINAHL Plus, and GreenFile databases, examined the association between road traffic noise exposure and dementia or cognitive impairment. The review found associations between long-term traffic noise and mild cognitive impairment in several large cohort studies, with some evidence of a relationship with global cognitive scores. The reviewers noted substantial heterogeneity across study methods and outcomes — which is honest — but this is also characteristic of an evidence base still assembling its methodology rather than one that has reached a null finding. The pattern is consistent, not conclusive. In an area where the consequences of waiting for certainty are irreversible cognitive decline in large populations, the evidential threshold for precautionary action should not require the same standard as a pharmaceutical approval.

Animal studies have been clearer. Environmental noise exposure in rodent models has been shown to influence cognitive performance alongside changes in tau protein and beta-amyloid deposition — the pathological hallmarks of Alzheimer’s disease. This does not prove causation in humans. It establishes a biologically plausible mechanism of sufficient seriousness to warrant longitudinal human investigation at scale. That investigation is now underway in several European cohort studies. Results will accumulate over the next decade.

What Remains Poorly Understood

The evidence base has real gaps. Most studies measure noise exposure through modelled traffic noise at residential addresses, not through personal exposure monitoring across daily environments. Workplaces, commutes, and social settings contribute exposure that residential modelling cannot capture. The dose-response relationship below 50 decibels — the level at which most regulatory standards begin to relax — is not well characterised. And the interaction between noise exposure and other environmental stressors (air pollution, heat, light at night) is almost entirely unexamined in large population studies.

These are research gaps, not reasons for inaction. The evidence that exists — on cardiovascular disease, on sleep disruption, on cognitive function in older adults — is already sufficient to justify treating noise exposure with the same systematic seriousness that urban air quality policy received once the particulate matter evidence became undeniable. The difference is that air is invisible and noise is audible. It has been easier to treat audibility as merely personal — a complaint rather than a condition.

The evidence that noise is a modifiable risk factor for dementia is not yet conclusive. But in populations that are already ageing faster than health systems can absorb, the cost of waiting for certainty is measured in years of functional life.

What the evidence now supports, at minimum: that chronic ambient noise exposure is a cardiovascular risk factor of moderate effect size at exposure levels that describe hundreds of millions of people’s daily environments; that it disrupts the sleep architecture whose importance for cognitive health and metabolic regulation has become one of the clearest findings in ageing research; that it operates through stress pathways established independently in the cardiovascular and neurological literature; and that its burden is distributed by the same geography that distributes other environmental health inequalities.

That is enough to ask why it is not in the clinical risk assessment. The answer, in all likelihood, is that there is nothing to prescribe. Risk factor identification in medicine has tended to follow treatment availability. Noise has no pill. Its remediation requires urban infrastructure policy, building standards, transport planning, and zoning — the apparatus of governance, not pharmacology. Medicine is not well structured to advocate for its patients in those arenas. But it could at least begin by naming the exposure.