AudioNotch Went Through a Randomized Controlled Trial. Here's What Happened.

In 2019, researchers at Northwell Health started a randomized controlled trial with AudioNotch as the intervention. Sixty-five tinnitus patients were randomly assigned to receive AudioNotch or standard care and followed for a full year. The updated 12-month analysis was published in Otology & Neurotology in 2026.

We did not run this study, fund it, or have any say in it. We found out the full results the way everyone else did: by reading the journal. This post explains what the researchers found, what the numbers mean, and, because this matters just as much, what the study does not show. If you want the compact version with the charts, it lives on our research page.

The short version

• Patients randomly assigned to AudioNotch improved significantly more over 12 months than patients assigned to standard care (Tinnitus Handicap Inventory regression: β −10.02, 95% CI −18.20 to −1.83, P=0.017).
• At 6 weeks, 70% of the AudioNotch group was still using sound therapy versus 32% of the standard care group (P=0.020). That is a 2.2x difference in adherence.
• Hours of reported use did not predict improvement (P=0.72). The researchers credit removing the cost barrier plus psychological factors: empowerment, perceived support, and accountability.

Who ran it and how

The trial was run by the Department of Otolaryngology at Northwell Health, with approval from its Institutional Review Board (IRB 25-0171). Enrollment ran from September 2019 to May 2023 at a tertiary referral center.

Adults with subjective tinnitus were randomized into two groups. The treatment group received a free subscription to AudioNotch. The standard of care group received usual management and could use notched sound therapy if they chose, but had to pay for it themselves. Everyone completed the Tinnitus Handicap Inventory (THI), a validated 25-question measure of how much tinnitus interferes with daily life, at enrollment and again at 6 weeks, 3 months, 6 months, and 12 months.

A preliminary analysis of the same trial was published in early 2023, covering the first 3 months. At that point there was no significant between-group difference in THI change, and we are linking it here because pretending it does not exist would be exactly the kind of selective citation this field has too much of: PMID 36624601. The 2026 paper is the updated, expanded analysis over the full 12 months, and it reaches a different conclusion because it has more patients and four times the follow-up.

What they found

The headline result is the regression across the full year: assignment to the AudioNotch group was associated with a 10-point greater reduction in THI than assignment to standard care (β −10.02, 95% CI −18.20 to −1.83, P=0.017). For context, the THI runs 0 to 100 and a 7-point change is commonly treated as the minimal clinically important difference, so a 10-point between-group difference is a meaningful gap, not a statistical curiosity.

By the 12-month visit, the AudioNotch group's mean THI change was −9.6. The standard care group had moved in the other direction, +5.0. Patients in the AudioNotch arm were also far more likely to still be participating at all: 93.8% of the standard care group had been lost to follow-up by 12 months, versus 69.7% of the AudioNotch group.

The adherence result is just as interesting. At 6 weeks, 70% of patients given AudioNotch were using sound therapy versus 32% of controls (P=0.020). Patients given free access were 2.2 times as likely to use sound therapy at that visit. Baseline THI, age, tinnitus frequency, and tinnitus loudness did not predict who used the therapy, so this was not simply the worst-affected patients trying harder.

What the study does not show

Here is the part most companies would bury. Reported hours of notched sound therapy use were not associated with THI improvement (β 1.50, 95% CI −6.89 to 9.88, P=0.72). Patients who listened more did not reliably improve more.

That means this trial demonstrates that being given AudioNotch helped, and it does not demonstrate that the notch mechanism itself is what helped. The authors are explicit about their interpretation: providing a free service with structured follow-up likely made patients feel supported and gave them an active role in their own care, and those psychological mechanisms, empowerment, perceived support, accountability, can meaningfully change how handicapping tinnitus feels. Their conclusion to clinicians is that providers should consider offering inexpensive or free sound therapies with brief structured follow-up.

Other limitations the paper reports: attrition was high at the later visits, very few patients hit the prescribed 2 hours per day of listening, use was self-reported, and the final sample of 33 and 32 per group fell just short of the 38 per group the power analysis called for. Trials of this size can miss real effects and can overstate fragile ones. It is one trial, not a settled question.

Why this result still matters

Tinnitus has no cure, and the marketing around it is notoriously ahead of the evidence. Against that backdrop, the bar we think consumer products should meet is simple: independent researchers, random assignment, a validated outcome, published in a peer-reviewed journal, warts included. This trial meets that bar, and AudioNotch was the app on the table.

The result supports keeping the product easy to start and inexpensive: the frequency tuner is free, and website purchases come with a 30-day money-back guarantee because results vary from person to person. The trial suggests that lowering cost barriers and providing structured follow-up can affect outcomes even when reported hours of use do not.

How it fits the wider evidence

The Northwell trial tested AudioNotch as a product. The notched sound method itself has a separate research record, and it is honestly mixed: Okamoto et al. (2010, PNAS) and Teismann et al. (2011, PLOS ONE) found reductions in tinnitus loudness and tinnitus-related cortical activity; Stein et al. (2016), a 100-participant trial of notched music, was null on its primary outcomes; Tong et al. (2023, Ear and Hearing) found tailor-made notched music training effective with better compliance than tinnitus retraining therapy; and meta-analyses disagree with each other. We keep a running, sourced summary in our ranking of sound therapies and on the research page.

The citation trail

Sturm H, Adamovich-Zeitlin R, Pelosi S, Mulloly V, Svrakic M. The Role of Compliance and Cost in Regulating the Effectiveness of Notched Sound Therapy Phone Applications for Tinnitus Relief: Updated Analysis of a Prospective Randomized Control Trial. Otology & Neurotology. 2026;47(2):e164-e169. doi:10.1097/MAO.0000000000004730. PubMed. An author copy is freely readable on ResearchGate.

Disclosure, so nobody has to guess: AudioNotch provided free study subscriptions for the treatment arm and had no role in the study's design, analysis, or publication. Northwell Health and the authors do not endorse AudioNotch. Nothing in this post is medical advice; if you have tinnitus, talk to a doctor first.

If you want to see what the participants used, start where they started: find your tinnitus frequency with the free tuner.