No treatment reliably silences tinnitus. The evidence is about distress, not loudness.

There is no treatment with good evidence that it makes tinnitus stop. The interventions that do have evidence — cognitive behavioural therapy, hearing aids, sound therapy, tinnitus retraining therapy — work on how much the sound intrudes on a person's life, and they are measured on questionnaires about distress rather than on the loudness of the sound itself [s2] [s3]. The most rigorously reviewed of them, CBT, reduced tinnitus impact by the equivalent of 10.91 points on a 100-point scale against no treatment, on low-certainty evidence [s2].

What tinnitus is treated as

The American Academy of Otolaryngology-Head and Neck Surgery Foundation's clinical practice guideline defines tinnitus as the perception of sound without an external source, and reports that more than 50 million people in the United States have experienced it, an estimated prevalence of 10% to 15% in adults [s1]. A 2025 umbrella review put the worldwide figure at over 740 million [s3].

The guideline's scope is the first thing worth noticing: it addresses tinnitus that is both bothersome and persistent, lasting six months or longer [s1]. Its very first strong recommendation is that clinicians distinguish patients with bothersome tinnitus from those with non-bothersome tinnitus [s1]. The clinical target is the burden, not the percept.

Two of the guideline's other strong positions cut against common expectations. It makes a strong recommendation against imaging the head and neck in patients whose tinnitus does not localise to one ear, is non-pulsatile, and is not associated with focal neurological abnormalities or asymmetric hearing loss [s1]. And it recommends a prompt, comprehensive audiologic examination for tinnitus that is unilateral, persistent, or associated with hearing difficulties [s1] — because tinnitus frequently travels with hearing loss, and the hearing loss is the treatable part.

What has evidence behind it

Cochrane's review of CBT for tinnitus included 28 studies with 2,733 participants, all with tinnitus for at least three months, with mean ages from 43 to 70 and CBT courses running 3 to 22 weeks, mostly in hospitals or online [s2].

Against no intervention or a waiting list, CBT reduced the impact of tinnitus on quality of life with a standardised mean difference of −0.56 (95% CI −0.83 to −0.30; 10 studies, 537 participants), rated low certainty [s2]. Re-expressed on the Tinnitus Handicap Inventory, that is 10.91 points lower — against an estimated minimal clinically important difference for that scale of 7 points [s2].

The comparisons that matter more clinically are less flattering. Against audiological care, CBT reduced THI by 5.65 points (95% CI −9.79 to −1.50; 3 studies, 444 participants), rated moderate certainty — statistically significant, and below the 7-point threshold the review itself names as the minimum a patient would notice [s2]. Against other active controls such as relaxation, information provision or internet discussion forums, the difference was 5.84 THI points (SMD −0.30, 95% CI −0.55 to −0.05; 12 studies, 966 participants), also below that threshold [s2]. The single comparison against tinnitus retraining therapy, from one study of 42 participants, gave 15.79 points in CBT's favour with a confidence interval running from 3.67 to 27.91 — low certainty, and one small trial [s2].

On safety, seven studies either reported serious adverse effects or confirmed their absence through personal communication with the review authors: six reported none, and in one study a single participant in the CBT group worsened [s2]. CBT also produced a slight reduction in depression (SMD −0.34, 95% CI −0.60 to −0.08; 8 studies, 502 participants; low certainty), while the evidence on anxiety and general health-related quality of life was too uncertain to call [s2].

The 2025 umbrella review, which synthesised 44 systematic reviews across seven intervention domains through April 2025, reached a compatible picture: CBT, hearing aids, tinnitus retraining therapy and sound or music therapy consistently improved tinnitus-related outcomes, with THI reductions up to 14.50 points [s3]. Cochlear implantation produced the largest effects of anything reviewed (THI −29.97) but is reserved for people with moderate-to-profound sensorineural hearing loss not adequately helped by hearing aids [s3]. Neuromodulation and acupuncture showed modest or inconsistent benefits with high heterogeneity [s3].

What has been tested and does not work

The guideline recommends against a specific list for routine treatment of persistent, bothersome tinnitus: antidepressants, anticonvulsants, anxiolytics and intratympanic medications; Ginkgo biloba, melatonin, zinc and other dietary supplements; and transcranial magnetic stimulation [s1]. It offers no recommendation either way on acupuncture [s1].

The supplement market has not improved since. A 2026 systematic review of over-the-counter oral treatments identified nine randomised controlled trials covering 390 adults in total, testing melatonin, Ginkgo biloba, alpha-lipoic acid, açaí extract and several combination antioxidant or herbal formulations [s4]. The authors could not perform a meta-analysis because the trials were too heterogeneous and reported variance data too incompletely [s4]. Several reported within-group reductions in THI scores on active treatment, but findings were inconsistent across agents, placebo responses were common, and most products had been evaluated in only a single small study [s4]. Their conclusion is that no firm conclusion can be drawn about whether any OTC therapy provides symptomatic benefit [s4].

Nine trials and 390 people, spread across at least five different substances, is not a body of evidence that a product is ineffective. It is an absence of evidence that it works — which is the condition most tinnitus supplements are sold in.

The realistic version

Persistent bothersome tinnitus has treatments that reduce distress by amounts that are real but often modest, and are best documented against doing nothing rather than against other active care [s2]. Where hearing loss is present, the guideline recommends a hearing aid evaluation [s1], and hearing aids appear among the consistently beneficial interventions in the umbrella review [s3]. Where distress is the dominant problem, CBT has the strongest evidence of any behavioural approach [s1] [s2] [s3]. Nothing in the reviewed literature abolishes the sound.

This article is informational and is not medical advice. Tinnitus in one ear only, pulsatile tinnitus, or tinnitus with sudden hearing loss, neurological symptoms or dizziness are reasons for prompt clinical assessment.