The 2014 AAO-HNS tinnitus guideline strongly recommends separating bothersome from non-bothersome tinnitus and not scanning tinnitus without warning features; it also recommends CBT and a hearing aid evaluation, and recommends against drugs, supplements and TMS.
The American Academy of Otolaryngology–Head and Neck Surgery Foundation published Clinical Practice Guideline: Tinnitus in October 2014. At the time, its authors wrote, there were no evidence-based, multidisciplinary clinical practice guidelines to help clinicians manage tinnitus. The Academy's guideline page mentions no revision since, and when the VA/DoD work group summarised its own guideline in a journal in 2025, it described the 2014 document as the most recent US clinical practice guideline for tinnitus.
Who it is for
The guideline is written for any clinician who manages tinnitus, including non-physicians, and it concerns a specific group of patients:
- Adults, aged 18 and over.
- Primary tinnitus: in the Academy's plain language summary, tinnitus with no known cause, which may or may not come with hearing loss. Secondary tinnitus has a specific cause, such as impacted earwax or Ménière's disease.
- Tinnitus that is persistent and bothersome, persistent meaning it has lasted 6 months or longer.
The treatment statements are about persistent, bothersome tinnitus rather than the first weeks of a new sound, and the plain language summary notes that tinnitus may improve on its own, especially when it is mild and has lasted less than 6 months. Does tinnitus go away? covers what the long-term data show.
The guideline puts US prevalence at 10% to 15% of adults, more than 50 million people, and the Academy's fact sheet adds that about 20% of adults who experience tinnitus will need clinical intervention. Its authors included otolaryngologists, an audiologist, a psychiatrist and consumer advocates, among others. Its literature searches ran up to April 2013.
How to read its grades
A 2023 systematic review of ten tinnitus guidelines records that the US guideline used the American Academy of Pediatrics' classification of recommendations. That scheme judges the overall quality of the evidence, then the expected balance of benefit and harm, and from those sets one of four strengths: strong recommendation, recommendation, option or no recommendation. A statement against something is graded on the same scale.
The Academy's fact sheet turns the grades into verbs. Clinicians should do what is recommended, may do what is an option, and should not do what is recommended against. On acupuncture it says no recommendation can be made.
One counting quirk: the executive summary speaks of 13 recommendations and the fact sheet numbers them 1 to 13, but statement 2 comes in two parts at two different strengths, so there are 14 graded statements in all. The numbers below are the fact sheet's.
The two strong statements
Only two statements carry the top grade, and neither is about treatment.
- Statement 4, strong recommendation: tell bothersome tinnitus from non-bothersome tinnitus. Everything the guideline says about treatment applies to the bothersome group. How tinnitus is measured covers the questionnaires used to judge this.
- Statement 3, strong recommendation against: do not scan tinnitus without warning features. Clinicians should not order imaging of the head and neck for tinnitus that does not localise to one ear, does not pulse, and comes with no focal neurological abnormality and no asymmetric hearing loss. The fact sheet gives the risk of radiation exposure as part of the reason. Do you need a scan for tinnitus? covers the other side of that line: one-sided tinnitus, pulsatile tinnitus, neurological signs and uneven hearing are the features that take a case outside it.
What it says clinicians should do
Six statements are graded recommendation:
- Statement 1: a targeted history and physical examination at the first visit, to find conditions that, identified and managed promptly, may relieve the tinnitus. The plain language summary gives earwax blocking the ear canal and fluid behind the eardrum as examples.
- Statement 2A: a prompt, comprehensive hearing test when tinnitus is in one ear, has persisted for 6 months or more, or comes with hearing difficulty. Your first audiology appointment describes what that involves.
- Statement 5: separate recent-onset from persistent tinnitus, to prioritise treatment and frame conversations about natural history and follow-up.
- Statement 6: education about management strategies for people with persistent, bothersome tinnitus.
- Statement 7: a hearing aid evaluation where persistent, bothersome tinnitus comes with documented hearing loss. The wording matters: it recommends an evaluation, not a device for everyone. Hearing aids for tinnitus covers who tends to benefit.
- Statement 9: cognitive behavioural therapy for persistent, bothersome tinnitus. CBT for tinnitus covers the trials.
What it leaves as options
Two statements are options, things clinicians may do:
- Statement 2B: a comprehensive hearing test at the first evaluation for anyone with tinnitus, regardless of side, duration or how well the person thinks they hear.
- Statement 8: sound therapy for persistent, bothersome tinnitus. The plain language summary suggests everyday devices, such as smartphones and radios, for delivering it. Sound therapy for tinnitus covers the evidence, and the sound library is a free place to start.
An option is not a rejection. It sits a tier below a recommendation, which in this scheme means a weaker case on the evidence or on the balance of benefit and harm.
What it recommends against
Three statements are graded recommendation against, each for persistent, bothersome tinnitus:
- Statement 10: antidepressants, anticonvulsants, anxiolytics or intratympanic medications for routine treatment. The qualifier is routine treatment of the tinnitus; drugs for tinnitus covers where medication does have a role. Intratympanic steroid injections have their own article, and gabapentin, the anticonvulsant most often trialled, has an encyclopedia entry.
- Statement 11: ginkgo biloba, melatonin, zinc or other dietary supplements. Each named product has its own page: ginkgo, melatonin, and magnesium and zinc.
- Statement 13: transcranial magnetic stimulation for routine treatment. TMS for tinnitus covers the trials.
Where it declined to rule
Statement 12 makes no recommendation on acupuncture. The plain language summary says there is not enough evidence to recommend or discourage it. That is a verdict on the trials, not a finding that acupuncture fails, and acupuncture for tinnitus covers why the question is still open.
What the evidence has done since
A 2016 analysis, whose authors included the guideline's methodologist, took the randomised trials the guideline's own search had found, 147 of them, and asked who they applied to. Nearly all took place in specialist settings, more than half did not define tinnitus, the median follow-up was 3 months, and only 20% were at low risk of bias. Its authors concluded that the trials apply best to older adults with long-standing tinnitus seen in specialist settings.
- CBT has held up, over the short term. A 2020 Cochrane review of 28 studies and 2,733 participants found CBT may reduce the impact of tinnitus on quality of life: 10.91 points lower on the Tinnitus Handicap Inventory than a waiting list, against a 7-point threshold for a meaningful change, on low-certainty evidence, and 5.65 points lower than audiological care on moderate-certainty evidence. There was no evidence at 6 or 12 months. The 2023 review of ten guidelines found counselling and CBT were the only treatments recommended by every guideline that addressed them.
- Sound therapy is still an honest option. The 2018 Cochrane review of sound therapy found 8 studies with 590 participants and no data at all for its main comparisons, so no evidence that devices beat a waiting list, a placebo or education alone. A 2026 umbrella review of 44 systematic reviews reads more warmly, listing sound and music therapy among the approaches that consistently improved tinnitus outcomes, alongside CBT, hearing aids and tinnitus retraining therapy.
- Hearing aids appear in that same list, and the 2024 VA/DoD guideline suggests them for tinnitus management in adults with hearing loss.
- The newest data back the TMS call. A 2025 meta-analysis of 16 randomised trials and 1,105 patients found rTMS beat sham on the Tinnitus Handicap Inventory at one month but not at six, with no significant effect on the Tinnitus Questionnaire or loudness matching. A 2026 meta-analysis of 8 randomised trials of rTMS, with 432 participants, found no significant effect (SMD −0.15, 95% CI −0.37 to 0.07), with no heterogeneity between trials.
- Supplements are where guidelines now split. The 2022 Cochrane review of ginkgo (12 studies, 1,915 participants) found it may have little to no effect on tinnitus severity against placebo: a difference of 1.35 points on the Tinnitus Handicap Inventory in the two trials that could be pooled, on very low-certainty evidence. A 2025 network meta-analysis of 60 trials ranked ginkgo with vitamins first for severity, but only 22% of its trials were at low risk of bias and its severity comparisons carried very low-certainty evidence. A 2026 Korean guideline now says ginkgo may be considered, while strongly recommending against routine zinc and vitamin supplements. The ginkgo article traces where that disagreement comes from.
- Nothing strong enough to overturn the position on drugs and injections has appeared. A 2022 meta-analysis of four double-blind, placebo-controlled trials found no evidence that intratympanic dexamethasone improved tinnitus more than placebo (odds ratio 1.38, 95% CI 0.53 to 3.61), although seven uncontrolled retrospective studies had reported improvement in 35.9% to 91.3% of patients. The 2025 network meta-analysis above ranked acamprosate first for annoyance and fluoxetine, an antidepressant, first for loudness, on the same weak trial base. The 2024 VA/DoD guideline suggests against a longer list: anticonvulsants, antidepressants, antiemetics, antithrombotics, betahistine, intratympanic corticosteroid injections and NMDA receptor antagonists.
- Acupuncture is still unresolved. The 2026 umbrella review found modest or inconsistent benefits, with high heterogeneity, for acupuncture and neuromodulation alike, and the VA/DoD guideline found insufficient evidence to recommend for or against it.
How it compares with the newer guidelines
The 2024 VA/DoD guideline reviewed evidence available up to April 2023 and made 25 recommendations, none of them strong: its work group said that, given rigorous adherence to GRADE and a lack of randomised trials, it did not have the evidence to make a strong one. Side by side:
- Where they agree: CBT (AAO-HNS recommends it; VA/DoD suggests it), hearing aids where there is hearing loss, and against ginkgo and other supplements, anticonvulsants, antidepressants and intratympanic medication.
- Sound therapy: an option for AAO-HNS; VA/DoD suggests the therapeutic use of sound for self-care.
- TMS: AAO-HNS recommends against it; VA/DoD finds insufficient evidence for or against. The VA/DoD review closed before the 2025 and 2026 meta-analyses above were published.
- Acupuncture: no recommendation in either, in two vocabularies.
- Strength: the 2014 document has two strong statements, both about assessment; VA/DoD has none.
None of the 2014 statements addresses cochlear implants, tDCS, low-level laser therapy, app-based self-management or questionnaires for monitoring treatment, all of which VA/DoD covers. Neither US guideline addresses bimodal neuromodulation devices.
The 2026 Korean guideline, on evidence it describes as mostly low or very low certainty, departs further: it conditionally recommends tDCS or rTMS as an add-on treatment and says ginkgo may be considered. The UK's 2020 guideline has its own walkthrough, and seven tinnitus guidelines compared sets all the major ones side by side.
How to use it
- The two strong statements are about assessment. Expect to be asked how much the tinnitus bothers you, and do not expect a scan unless it has one of the features above.
- Ask for a hearing test. A prompt one is recommended when tinnitus is one-sided, has persisted 6 months or more or comes with hearing difficulty, and one is an option for anyone. If it finds hearing loss and the tinnitus is persistent and bothersome, the guideline recommends a hearing aid evaluation.
- If you are offered something on the against list (a supplement, an antidepressant or anticonvulsant aimed at the tinnitus itself, an injection through the eardrum, or TMS), it is fair to ask why your case differs.
- Tinnitus in one ear, tinnitus that pulses, a new neurological symptom or hearing that is worse on one side needs a clinician's assessment; the when to seek care checklist sorts which signs are urgent. How tinnitus is diagnosed describes a full work-up, and the treatment comparison shows the AAO-HNS position beside each treatment.