Tinnitus: what is actually happening, and what genuinely helps

Tinnitus — the perception of sound without an external source — affects roughly 10 to 15 per cent of adults, with prevalence rising with age. For most it is intrusive occasionally. For perhaps one to two per cent it is severely distressing, and that distress correlates poorly with how loud the sound actually is.

Where the sound comes from

Not the ear, in most cases. The current model is that tinnitus arises from central changes following reduced auditory input.

When hearing is lost at particular frequencies — from age, noise exposure, or both — the auditory cortex receives less input in that frequency range. Neurons there increase their spontaneous firing and become more synchronised, a phenomenon sometimes described as central gain, analogous to a hearing aid turning up its own amplification and amplifying its own noise floor. The brain generates activity in a channel that has gone quiet, and that activity is perceived as sound.

This explains several observations: that tinnitus pitch usually corresponds to the region of hearing loss; that it can persist after the auditory nerve is severed; and that hearing aids, by restoring input, often reduce it.

Why distress and loudness come apart

Once the sound exists, what determines suffering is largely how the brain handles it — the limbic and attentional systems rather than the auditory ones.

People with severe tinnitus distress typically show a pattern: heightened attention to the sound, catastrophic interpretation of what it means, anxiety about it becoming permanent or worsening, sleep disruption, and avoidance of quiet. Each of those amplifies the others. The sound becomes salient in the way a dripping tap becomes unbearable at night while a much louder fan does not.

This is why the effective treatments are the ones that address attention and interpretation, and why "there is nothing that can be done" is both technically true about the sound and clinically wrong about the condition.

What has evidence

Cognitive behavioural therapy has the strongest evidence base of any tinnitus intervention. Cochrane and subsequent reviews consistently find it improves quality of life and reduces tinnitus-related distress. It does not reduce the loudness of the sound, and that is not the target — the target is the relationship with it. Internet-delivered CBT for tinnitus has also shown benefit, which matters for access.

Hearing aids, where there is hearing loss. Restoring auditory input reduces central gain, and the ambient sound they amplify partially masks the tinnitus. Given how commonly tinnitus and hearing loss coexist, this is often the first practical step and it is under-used.

Sound therapy. Low-level background sound — a fan, a radio at low volume, environmental sound generators, or a bedside sound machine — reduces the contrast between tinnitus and silence. The evidence is modest but the intervention is harmless and cheap, and it helps most with sleep, which is where the burden often concentrates.

Tinnitus retraining therapy combines counselling with sound therapy over a long period. Evidence is mixed, and the counselling component may be doing most of the work.

Mindfulness-based approaches have accumulated reasonable trial evidence, comparable to CBT in some studies.

What does not work

Worth stating clearly, because a great deal is sold.

Ginkgo biloba has been repeatedly tested and does not work. Zinc, melatonin and various vitamin preparations lack convincing evidence — melatonin may help sleep, which helps indirectly. There is no drug licensed to treat tinnitus itself. Antidepressants and anxiolytics may be appropriate where there is comorbid depression or anxiety, which is common, but they do not treat tinnitus.

Repetitive transcranial magnetic stimulation has been studied extensively with disappointing and inconsistent results. Bimodal stimulation devices, combining sound with electrical stimulation of the tongue or neck, have produced encouraging trial results and are commercially available in some countries; the evidence is promising rather than settled.

Practical management

Get your hearing tested. The single most useful first step.

Avoid silence. Quiet rooms make tinnitus louder by removing competing sound. Low-level background noise, particularly at night.

Protect your hearing from further noise exposure, which can worsen it.

Address sleep deliberately. The relationship is bidirectional, and poor sleep makes tinnitus more intrusive the following day.

Reduce monitoring. Checking whether it is still there guarantees noticing it. This is one of the specific behaviours CBT targets.

Caffeine is widely believed to worsen tinnitus; controlled studies have generally not supported this, and abrupt caffeine withdrawal can itself cause symptoms.

The realistic prognosis

For most people tinnitus does not disappear, and it does not need to. Habituation — the process by which the brain stops flagging a constant, unimportant stimulus — happens naturally over months to years in the majority, and it is what the effective treatments accelerate.

The people who do worst are those who are told nothing can be done and are left to manage the fear alone. The condition responds poorly to treatment aimed at the sound and quite well to treatment aimed at everything else.