Considerations for fitting, programming, and verification of sound therapy delivered by hearing aids

Abstract

Objective:

Identification of fitting and programming considerations for sound therapy in hearing aids based on electroacoustic characterization across brands, including assessment of interactions between sound therapy and other digital signal processing (DSP) features.

Methods:

The study was conducted in two parts: (1) electroacoustic characterization of default sound therapy settings, and (2) analysis of interactions between the sound generator, feedback cancellation, and noise reduction features. Five receiver-in-the-ear hearing aids from different manufacturers were assessed using the Ahead Simulations Clinical Assistant for Research and Learning (CARL) manikin and the Audioscan Verifit 2 hearing aid analyzer. Devices were programmed for two standard audiograms using both dome and custom earmold couplings.

Results:

Default sound therapy outputs varied significantly across brands, influenced primarily by audiogram and coupling type. Feedback testing revealed that enabling sound therapy caused devices to exceed their stable gain limits, resulting in audible feedback. Noise reduction algorithms reduced sound therapy signals in open-fit configurations.

Conclusions:

Sound therapy features differ across hearing aids and can interact with other DSP functions counterproductively. These interactions are measurable using standard clinical verification tools, highlighting the importance of verification of hearing aid programs. Five clinical fitting practices were developed to address these considerations.

1 Introduction

Chronic subjective tinnitus is a condition in which individuals experience the sensation of sound without an external source, and which can vary in degree of intrusiveness (Cima et al., 2019). Although the pathophysiology of tinnitus is still disputed, tinnitus is thought to involve altered auditory or somatosensory inputs, and abnormal activity in central nervous structures (Langguth et al., 2013; Schlee et al., 2024). Sound therapy is considered a potential treatment for individuals with chronic subjective tinnitus, including those with concurrent hearing loss (Henry et al., 2008; Jastreboff and Jastreboff, 2000; McNeill et al., 2012). Sound therapy can include providing sounds to the patient that either reduce the contrast of the tinnitus, distract attention from tinnitus, or are subjectively soothing (Henry, 2022). In modern hearing aids, these may be available within the hearing aid itself or through connected software, reducing the need to provide sound therapy through specialized devices. The sound therapy approaches may aim to decrease perception of tinnitus, promote patient control and emotional regulation, enable habituation to tinnitus, and/or contribute to changes in the tinnitus neural pathways (Newman et al., 2008). Sound therapy can be delivered in different forms and has been shown to provide effective intervention for tinnitus (Kutyba et al., 2022; Liu et al., 2021; Tyler et al., 2020).

Tinnitus is frequently associated with sensorineural hearing loss and steeply sloping audiograms (Chung et al., 1984; König et al., 2006); as such, the use of hearing aids as an external sound generator is common in audiological care for tinnitus. Alongside standard amplification, modern hearing aids often include an onboard sound generator that can be programmed to deliver broadband noise as a background sound for tinnitus masking—one form of sound therapy for tinnitus intervention (dos Santos et al., 2014; Lelic et al., 2024). This feature is increasingly available in modern hearing aid technology, with devices that have the capability to provide both environmental amplification and sound therapy being referred to as combination devices. Recommended practice includes the individualized fitting and verification of hearing aids to meet patients' needs (APSO., 2021). However, the availability, options, and customization capabilities of these sound-generating features vary across manufacturers, and this variability has not been well characterized. To address this, the primary aims of this study are to (1) investigate the electroacoustic characterizations of default sound therapy settings across different hearing aid brands, hearing loss severity, and coupling types, and (2) develop a concise and actionable set of considerations for fitting, programming, and verification of hearing aid tinnitus sound therapy (Boisvert et al., 2017; Moodie et al., 2011).

The purpose of the following investigations is to address two primary aims: Aim 1: Investigating between-product variability in sound therapy and interactions with digital signal processing features. Variability in the implementation of sound therapy has been demonstrated by Hong et al. (2020), who examined the white-noise characteristics across three different hearing aids. The researchers found notable differences in both intensity and frequency response across brands. They also observed that software settings for broadband noise often did not match the hearing aid's actual output as measured by verification in an Audioscan Verifit 2 hearing instrument test box (Hong et al., 2020). This study was limited in scale to a small selection of hearing aids with dome coupling and a single hearing loss configuration, leaving unanswered questions about how default sound settings vary across different sloping hearing loss profiles and coupling types. Additionally, there is little research on how broadband noise–based sound therapy interacts with other digital signal processing (DSP) features of hearing aids. Manufacturer instructional guides either do not specify whether such interactions exist or outline other programming considerations, leaving clinicians to navigate these complexities independently. In a review article, Lansbergen and Dreschler (2020) developed a framework to objectively compare hearing aids regardless of brand or style. The framework details a set of essential hearing aid features that belong to domains such as signal processing, comfort, and adaptation (Lansbergen and Dreschler, 2020). Of these domains, sound therapy was not specifically categorized. As a result, we selected features from this framework and applied them in this study to assess their potential interactions with sound therapy, specifically noise reduction and an active feedback manager. Noise reduction (NR) was selected because we hypothesized that sound leakage from the ear canal during sound therapy may be captured by the hearing aid microphones and processed by the NR algorithm. As the sound therapy is a steady-state broadband noise, it is possible that basic NR algorithms are capable of reducing this type of signal (Ricketts et al., 2017). This interaction has not been well documented in the literature. Furthermore, as the ear canal sound pressure level is a major contributor to audible feedback, we hypothesized that the additive effects of sound therapy with environmental sound amplification may exceed stable gain limits (Ricketts et al., 2008). This warrants investigation, as audible feedback is a deterrent to consistent hearing aid use (Kochkin, 2007). The present study addressed the following research questions: (a) How do the default intensity levels of manufacturer-programmed sound therapy vary across coupling types and hearing loss configurations? (b) Does digital noise reduction reduce the output level of sound therapy in the ear canal? (c) Does sound therapy delivered adjacent to amplification exceed maximal stable gain and result in audible feedback? Aim 2: Adapting Existing Fitting Guidelines for Tinnitus Sound Therapy. Despite the current evidence that combination devices are effective for tinnitus management, it has been suggested that too few patients receive sound therapy, leaving unmet needs (Henry et al., 2021, 2022). One way to enhance consistent provision of care may be to further specify the clinical care pathway for providing sound therapy in hearing aids. There are several tinnitus management protocols, such as Progressive Tinnitus Management and Tinnitus Retraining Therapy, which outline the goals and general provision of sound therapy (Jastreboff and Hazell, 2004; Myers et al., 2014). However, these guidelines may not provide specific steps on how to combine hearing aid provision with sound therapy. In 2020, the British Society of Audiology published an audiological practice guide for fitting combination hearing aids for subjects with tinnitus; in this guide, the authors emphasize the device user's preferences, highlighting the individualized approach required for tinnitus management. As with other tinnitus management guidelines, the clinical procedures for fitting, programming, and verifying sound therapy are not specified (British Society of Audiology., 2020). This contrasts with hearing aid fitting guidelines, which typically recommend using prescriptive targets and electroacoustic verification to provide person-specific sound levels (APSO., 2021). The findings of this study should be interpreted within the context of hearing aid fittings for individuals with hearing loss and may not generalize to sound therapy delivered by hearing aids in individuals with normal hearing sensitivity.

2 Materials and methods

2.1 Study design

An electroacoustic investigation was conducted on commercially available hearing aids released from 2020 to 2023, using Ahead Simulations' Clinical Assistant for Research and Learning (CARL) basic manikin as a simulated ear (AHead Simulations., 2024; Folkeard et al., 2023). An Audioscan Verifit 2 hearing aid analyzer was used to collect on-ear measures and verification of settings (Audioscan., 2023).

2.2 Simulated hearing loss

Two standard audiograms were selected: N4, a moderate sloping to moderately-severe hearing loss; and S2, a normal hearing sensitivity sloping to profound hearing loss (Bisgaard et al., 2010). These hearing losses were selected because individuals with sloping hearing losses are likely to have tinnitus (Chung et al., 1984). These hearing losses also represent clinically challenging fittings due to their combined candidacy for both venting and high gain, which can create the potential for feedback. For a sloping hearing loss, a typical hearing aid fitting would provide greater gain in the high-frequency regions where hearing sensitivity is reduced. Because audible feedback is commonly associated with frequencies between 2 and 4 kHz, feedback-management features are incorporated into the fitting to minimize feedback while maintaining adequate high-frequency amplification. Furthermore, individuals with sloping losses have better sensitivity in low frequencies compared to high frequencies, larger venting options are typically selected to improve comfort and sound quality. However, larger vents may also allow excessive unwanted noise into the ear canal. As a result, noise reduction features are often incorporated into these fittings to help mitigate the unwanted noise. The left ear hearing aid was programmed to the N4 hearing loss, and the right ear hearing aid was programmed to the S2 hearing loss.

2.3 Hearing aid fittings

Receiver-in-the-canal hearing aids from five different manufacturers were selected for this electro-acoustic analysis using medium-powered receivers. The hearing aids were selected to represent independent technology offerings: Resound, Starkey, Oticon, Widex, and Phonak. Brand-specific program settings and hearing aid models are listed in Table 1; performance data will be anonymized for the remainder of this paper. Two coupling methods were also used for each hearing aid: an open-fitting with a single-walled dome and a closed-fitting with a custom earmold tip, made for each brand of hearing aid from ear impressions taken from the CARL manikin's ear (size large for the right, and bendy for the left).The different types of ears used on CARL are representative of the anatomical variability observed across individuals, both types are validated against a sample of human participants (Folkeard et al., 2023). Differences between ear canal anatomies is compensated with individually measured real ear coupler differences (RECD) for this study. The monaural hearing thresholds were entered into the Verifit system in dB HL, which were then converted to dB SPL (eardrum) using individualized manikin RECD values in combination with the appropriate reference equivalent threshold sound pressure level (RETSPL) values for insert earphones at each test frequency (Audioscan., 2026; Bagatto et al., 2005). The manikin was mounted on a swing arm that was positioned 50 cm from the Verifit 2 speakers, as per positioning recommendations (Audioscan., 2026). The position of the swing arm did not change between measures or when changing hearing aid to ensure uniformity of recording environment.

A single dome type was selected for use on all hearing aids as a partial control condition. Unlike custom earmolds, which are receiver-specific and would not universally fit each manufacturer's receiver type, the dome we selected fit every receiver reasonably. This approach was not ideal and could not be a complete control condition, as small deviations in the fit around the receiver could have caused slit leaks for sound to escape. This dome was selected to minimize extraneous variability associated with manufacturer-specific dome designs. Specifically, the selected dome (1) represents the average vented dome available across manufacturers, as compared to the REOIG measures from Blau et al. (2008) and average insertion loss from Winther Balling et al. (2019), (2) is a size and material that is compatible with the manikin ear and (3) which could be reliably and consistently fit to the same position inside the manikin ear. For each manufacturer's software, we selected the corresponding single-vented dome type. Although using a non-brand-specific dome may have altered the frequency response of some hearing aid receivers, amplification gain was verified to be clinically appropriate for a DSL v5 adult fitting. The differences observed across brands reflect the influence of coupling type itself on sound therapy and were therefore considered relevant to the study objectives. The earmold vent size was selected based on the manufacturer's software recommendations, which ranged from completely occluded to 3 mm vents. The openness of fit was measured by insertion gain, specifically insertion loss due to the attenuation of the coupling (Real ear occluded gain, REOG), calculated by comparing the Real Ear Occluded Response (REOR) to the Real Ear Unaided Response (REUR; REOG = REOG–REUG). All measurements were made using pink noise at 65 dB SPL. Measures were reviewed for accuracy in probe tube depth and acoustic seal prior to taking final measurements. On average, the custom earmolds reduced the gain from 250 to 8,000 Hz by 13.7 dB (range: −1.27 to −21.55 dB) dB for the S2 audiogram and 7.4 dB (range: −0.61 to −13.48 dB) for the N4 audiogram, exceeding the attenuation of the domes, which attenuated about 6.7 dB and 3.6 dB for the S2 and N4 fittings, respectively. Overall, the earmolds attenuated more sound compared to the dome except for in Brand D (Table 2), which was a more open fit than the dome, with average attenuation of 1.3 and 0.6 dB for S2 and N4 fittings, respectively. A contributing factor to Brand B appearing more open than the dome condition may be slit leakage created by the fit of the receiver within the earmold. A number of factors can influence the effective acoustic openness of the fit, including the insertion depth and positioning of the receiver within the ear canal, venting characteristics of the coupling, and the placement of the probe tube during measurement (Agnew, 1996). Collectively, these factors may have contributed to increased acoustic leakage and a more open REUR/REOR response in this condition.

All hearing aids were programmed to binaural DSL v5 adult prescriptive targets (Scollie et al., 2005). The ear canal acoustics of the CARL manikin were measured using the real-ear-to-coupler difference (RECD) with a foam tip, as incorporated into the prescription. The DSL prescription was selected as it was found to be preferred by hearing aid users with tinnitus compared to other prescriptive methods (Shekhawat et al., 2013; Shetty and Basavaraj, 2022). Hearing aid output was measured on-ear using the Audioscan Verifit 2, and the fit to target was verified to be within 5 dB root mean square error (RMSE; Dao et al., 2021; Waechter and Jönsson, 2022). Probe tube placement was verified to be within 5 mm of the manikin eardrum prior to each hearing aid fitting using the Probe Tube Guide feature on the Verifit 2.

2.4 Sound therapy programming

The hearing aids offer a variety of broadband noise stimuli, most of which are generated using Gaussian noise and can be spectrally shaped (e.g., white noise, pink noise, or speech-shaped noise). For this study, the noise stimulus was selected to approximate participants' hearing thresholds whenever this option was available within the manufacturer's fitting software. This feature had different names, such as “Audiogram shaped”, “Hearing loss shaped”, “High frequency noise”, or simply “Shaped”. This shaping was selected as Hong et al. (2020) had previously investigated white noise sound therapy, and to assess the effects on intensity levels in higher frequencies of the sound due to the sloping hearing loss. However, Brand E did not provide a hearing-threshold-shaped noise option; therefore, white noise was selected as the closest available alternative. Other than enabling these programs and selecting this custom shaping, no other parameters of the sound therapy were modified from the software's pre-selected settings, such as intensity and modulation. A list of software and programming parameters for sound therapy and digital sound processing features can be found in Table 1.

2.5 Electroacoustic measures of default sound therapy settings

The Speech-live feature on the Audioscan Verifit 2 was used to capture the frequency response of the default sound therapy. Speech-live is a measure that records the hearing aid output in the absence of an external signal for a duration of approximately 12 s (Audioscan., 2026). A baseline recording was made with the hearing aids set to the default sound therapy program with noise reduction disabled. The overall level of the sound therapy was then calculated from 1/3 octave bands using power-sum analysis from 200 to 8,000 Hz, in dB SPL (eardrum).

2.6 Noise reduction measures

To evaluate the potential interaction between NR algorithms and sound therapy, Speech-live was used to record the frequency response of the sound therapy under various noise reduction conditions. The first condition was sound therapy with the noise reduction setting disabled, followed by sound therapy with the noise reduction feature enabled at default settings. For the second condition, the NR was enabled prior to recording. This was to ensure that the algorithm had time to activate and reach an asymptotic performance before the Speech-live recording (Scollie et al., 2016). A timer was used to standardize this interval across recordings, at approximately 20 s per recording. The recordings were made multiple times for each hearing aid to ensure stability, but only the final recording was used for data analysis. Both power sum averaging and arithmetic averaging across spectral level differences. Overall levels were calculated for conditions from 1/3-octave bands. The calculations together offered a more comprehensive characterization of the effects of NR, capturing both overall level changes and frequency-specific effects. This difference between conditions represents the reduction of sound therapy caused by the NR algorithm in dB SPL (ear drum). This measurement was possible in all hearing aids except for Brand E, where the software prevented enabling noise reduction settings while sound therapy was active.

2.7 Active feedback manager measures

To assess active feedback manager capability, the maximum stable gain was measured for each hearing loss configuration and coupling type. Similar to the experiment by Marcrum et al. (2018), a baseline measure was first obtained without sound therapy (MSG _Baseline); the global gain was increased in 1 dB increments while a researcher monitored the hearing aid output in CARL's ear canal via headphones connected to the Verifit. To simulate a typical clinical listening check, researchers held their hands near the hearing aids, altering the feedback path, in order to detect sustained squealing while a 65 dB SPL speech passage played continuously from the Verifit. Once audible feedback was detected, the global gain was then reduced until feedback ceased, and the level was lowered an additional 2 dB to determine the maximal stable gain. The output was recorded using the same speech passage. In the follow-up measure, sound therapy was enabled (MSG_ST), and the same procedure as at baseline was performed, ensuring that only the amplification gain was manipulated and the sound therapy level remained constant. Maximum stable gain for 2,000, 3,000, and 4,000 Hz was calculated by subtracting Real-Ear Unaided Response from the Real-Ear Aided Response obtained (i.e., Real-ear Insertion Gain; Marcrum et al., 2018) for both conditions.

To quantify the effect of sound therapy on the presence of audible feedback, the Additional Gain Before Feedback (AGBF) measure was computed, following published procedures (Marcrum et al., 2018). The AGBF is simply a calculation of the difference between the maximum stable gain measured with the tinnitus sound therapy and the unadjusted hearing aid gain set to DSLv5 adult targets alongside sound therapy. Negative values indicate that amplification reductions were required to maintain feedback stability when sound therapy was enabled. The effect of sound therapy on available headroom was evaluated across the full frequency range, as the sound therapy stimuli contained low-frequency energy in addition to mid- and high-frequency components.

2.8 Analysis of results

Recall that the aims of the study was to explore the range of between-product variability of interactions between signal processing and tinnitus sound therapy, to better inform fitting guideline modifications. As such, a small illustrative sample of fully-verified fittings was collected. This sample does not support inferential statistics, so descriptive analyses of range of variation will be presented below. Where variations exceed commonly-accepted clinical tolerances, clinical strategies in the form of modified fitting protocols will be developed.

3 Results

3.1 Variability of default sound therapy settings

Default sound therapy levels varied across brands, depending on the coupling type and hearing-loss configuration. Figure 1 presents the 1/3-octave band output levels for each sound therapy stimulus. To facilitate comparison with audiometric thresholds, the frequency axis is labeled using the standard audiometric octave frequencies in accordance with recommendations from the (American Speech-Language-Hearing Association., 1990). Between-brand variability was substantial, with differences in default levels (lowest to highest) of 18.7 dB SPL (ear drum) for S2 dome coupling (Figure 1, panel A), 25.0 dB SPL (ear drum) for S2 earmold coupling (Figure 1, panel B), 22.1 dB SPL (ear drum) for dome-fit N4 (Figure 1, panel C), and 22.3 dB SPL (ear drum) for earmold-fit N4 (Figure 1, panel D).

Figure 1

Bisgaard et al., 2010). All values are expressed in dB sound pressure level (SPL) as measured in the ear of a manikin, using a commercially available hearing aid analyzer.

For the S2 audiogram, dome couplings yielded intensities between 59.8 and 78.5 dB SPL (eardrum; mean = 71.6 dB SPL, SD = 9.4), while earmold couplings ranged from 59.6 to 84.6 dB SPL (eardrum; mean = 70.8 dB SPL, SD = 11.3). For the N4 audiogram, dome couplings produced sound therapy intensities ranging from 69.2 to 91.3 dB SPL (eardrum; mean = 83.4, SD =8.8), whereas earmold couplings ranged from 65.4 to 88.1 dB SPL (eardrum; mean = 79.1 dB SPL, SD = 8.7).

Overall, the coupling that produced higher in-ear sound therapy levels depended on both brand and audiometric configuration. For the N4 audiogram, dome fittings yielded, on average, 4.3 dB SPL (eardrum) higher sound therapy levels than earmold fittings, with differences ranging from −1.3 to 14.1 dB SPL (eardrum; SD = 5.9). For S2 audiograms, results were mixed across brands, with an average difference between dome and earmold couplings of 0.8 dB SPL (eardrum), but a range of −6.1 to 10.3 dB SPL (eardrum; SD = 6.3).

3.2 Interactions between noise reduction and sound therapy

Differences between power-sum averaging and arithmetic averaging of spectral level differences were present when analyzing the effects of NR on sound therapy. Each approach has distinct advantages and limitations: power-sum averaging places greater emphasis on larger spectral differences and is more representative of perceived overall level changes, whereas arithmetic averaging is more sensitive to localized spectral differences but may be influenced by low-level spectral regions that contribute less to perceived loudness. Both metrics were presented in Figure 2. Positive values reflected a reduction in sound therapy output due to the NR algorithm, whereas negative values indicated an increase in sound therapy level, and likely minimal interaction between the noise reduction algorithm and sound therapy.

Figure 2

Examining the interaction through power-sum averaging revealed that Brands A and D were least affected, with less than 2 dB change in sound therapy levels between conditions and Brands B and C were observed to have varying effects on the sound therapy level, depending on both the audiogram and the coupling type. Across all brands, the N4 audiogram overall level changes ranged from −0.1 to 7.2 dB (mean = 1.6, SD = 3.2) for earmold fittings and from −0.1 to 9.4 dB (mean = 2.3, SD = 4.1) for dome fittings. For the S2 audiogram, changes ranged from −0.3 to 5.7 dB (mean = 1.7, SD = 2.6) for dome coupling and from 0.00 to 6.2 dB (mean = 1.7, SD = 2.7) for earmold fittings. In Brand B specifically, S2 audiograms were more susceptible to NR effects, with earmold coupling sound therapy reduced by 5.7 dB compared to dome coupling being reduced by 2.1 dB. N4 audiograms in Brand B were not as affected with a reduction in sound therapy of 0.8 and 2.2 dB for earmold and dome couplings, respectively. In Brand C, dome coupling types were most susceptible to NR effects regardless of audiogram, with reductions of up to 9.4 dB for N4 dome coupling and 6.2 dB S2 dome coupling, compared to 7.2 dB and 2.9 dB for N4 and S2 earmold coupling, respectively. Similarly, in the 1/3 octave band artimetic averaging, Brands A and D were least effected by the NR algorithm, with less than 2 dB of change between conditions. In general, Brand B sound therapy was reduced by the NR across all coupling types and audiograms, and for Brand C both audiogram and coupling type had varied effects of the NR. Across all brands, the N4 audiogram overall level changes ranged from −0.1 to 6.4 dB (mean = 1.8, SD =2.8) for earmold couplings, and 0 to 6.9 dB (mean = 1.9, SD = 3.0) for dome coupling. For the S2 audiogram, earmold coupling ranged from 0.6 to 4.6 dB (mean = 1.4, SD = 2.0), and −0.1 to 2.8 dB (mean = 1.0, SD = 1.5) for dome couplings. Brand B sound therapy was reduced by 2.5 dB on average across coupling types and audiograms. In Brand C, sound therapy was reduced by NR by an average of 6.7 dB for the N4 audiogram across dome and earmold coupling, and for the S2 audiogram, the sound therapy was reduced by an average of 3.7 dB across coupling types. Taken together, both the power-sum and arithmetic averaging approaches demonstrated that the impact of noise reduction on sound therapy output was highly dependent on hearing aid manufacturer, audiogram, and coupling configuration. While Brands A and D showed minimal changes in sound therapy levels, Brands B and C exhibited greater reductions, with the magnitude of these effects varying according to the measurement method use.

3.3 Maximum stable gain and additional gain before feedback with sound therapy

When sound therapy was enabled, audible feedback was present at DSL v5 adult prescriptive gain levels measured using a 65 dB SPL speech passage. The effect of sound therapy on feedback suppression capability was minimal. When pooled by coupling type across audiograms and frequencies (2,000–4,000 Hz), the mean difference in MSG_Baseline and MSG_ST was 1.6 dB (SD = 1.2) for dome coupling and 0.8 dB (SD = 1.1) for earmold coupling. Per frequency, changes were smaller for the N4 audiogram and larger for the S2 audiogram, with dome coupling showing relatively stable values across frequency (Figure 3). For the N4 audiogram, individual frequency differences ranged from −0.3 to 1.5 dB for dome coupling (Figure 3, panel A) and from 0.0 to 0.6 dB for earmold coupling (Figure 3, panel B). In contrast, the S2 audiogram exhibited larger changes, with dome coupling remaining relatively stable across frequency, ranging from 2.0 to 2.7 dB (Figure 3, panel C) and earmold coupling varying from 0.7 to 2.7 dB (Figure 3, panel D).

Figure 3

Additional Gain Before Feedback varied across devices with frequency and coupling condition. Across conditions, acoustic feedback occurred at DSL v5 adult gain levels adjacent to default sound therapy settings. In general, dome coupling provided greater AGBF than earmold coupling at low-to-mid frequencies (250–1,500 Hz), whereas differences between coupling conditions became more variable at higher frequencies, with several device configurations exceeding the MSG. In the low- to mid-frequency range (250–1,500 Hz), there was more headroom before the MSG was reached for dome coupling than for earmolds. N4 dome coupling had an average AGBF ranging from 7.5 to 11.1 dB (SD = 1.8), and 2.2–9.5 dB (SD = 3.1) for N4 earmold coupling. The S2 dome coupling had an average AGBF ranging from 3.5 to 11.4 dB (SD = 3.3), and the S2 earmold coupling had an average AGBF ranging from 0.6 to −0.5 dB (SD =1.1), which exceeded stable gain limits. For the N4 audiogram, dome coupling yielded minimal additional gain in the mid- to high-frequency range (2,000–4,000 Hz), with average AGBF values ranging from 2.3 to 1.1 dB before the MSG was reached. In contrast, earmold coupling demonstrated greater variability, with AGBF values spanning from 8.7 to −3.1 dB, indicating a gain reduction at higher frequencies where the MSG was exceeded. For the S2 audiogram, dome coupling produced AGBF values ranging from 8.4 to −1.9 dB across frequencies, while earmold coupling ranged from 5.8 to −3.3 dB, similarly reflecting reduced gain margins at higher frequencies. In the high frequencies (6,000–8,000 Hz), there was less available gain across conditions. Average AGBF for N4 dome coupling ranged from −4.0 to 2.8 dB (SD = 4.9). N4 earmold AGBF ranged from −6.3 to −2.1 dB (SD = 2.9). S2 dome coupling AGBF ranged from −4.0 to 2.2 dB (SD = 4.4), and earmold coupling ranged from 1.5 to 8.5 dB (SD = 4.9). A corresponding figure showing the per-brand and per-frequency AGBF values, along with the overall average, is provided in the Supplementary Material (Figure S1).

4 Discussion

In summary, we examined the electroacoustic characteristics of default sound therapy settings in receiver-in-the-ear hearing aids. The primary aim was to determine how these settings vary across manufacturers, as the lack of standardized guidelines may lead clinicians to rely on software-recommended presets during programming. Because each manufacturer uses its own sound therapy algorithms, these presets can differ substantially, contributing to variability in clinical practice. Additionally, the literature on electroacoustic verification of sound therapy signals is sparse, highlighting the need to investigate how verification can identify both the provided levels of sound therapy and any unwanted interactions between digital signal-processing features and tinnitus sound therapy outputs. Such interactions may complicate programming and troubleshooting, ultimately making the initial fitting of sound therapy less reliable and more variable. The results of this study revealed both between-brand variability in the levels of sound therapy provided by default, as well as interactions between the sound therapy and use of other signal processing features, summarized below.

4.1 Physical coupling of hearing aids for sound therapy and default settings

Electroacoustic characterization demonstrated that hearing aid coupling type influenced the intensity of default sound therapy, with the effect varying by brand and audiometric configuration. For the N4 audiogram, most brands showed higher in-ear sound therapy levels with dome couplings, whereas for the S2 audiogram, earmold couplings generally produced higher default sound therapy levels. These findings suggest that physical coupling should be considered not only for amplification requirements but also for its impact on sound therapy delivery.

The variability of default intensity settings across brands was evident, with large differences between the highest and lowest overall levels. This variability is important to understand as the intensity of sound therapy is related to therapeutic benefit (Hong et al., 2020). Currently, there are no intensity-related benchmarks for programming sound therapy delivered by hearing aids. However, a conceptually-related construct may be the Just Meaningful Difference (JMD) for Signal to Noise Ratio (SNR) has been determined (McShefferty et al., 2016). In this study, participants could detect SNR differences of 3 dB (i.e., the just-noticeable difference), but only SNR improvements exceeding 6 dB were judged meaningful enough to justify intervention. This measure bridges psychoacoustical performance and clinically important differences, which provides a useful point of reference for intervention selection and efficacy (McShefferty et al., 2016). Compared to this 6 dB JMD, the differences in tinnitus sound therapy levels across brands in this study were very large, with discrepancies of 18–24 dB SPL (ear drum) across frequencies, far exceeding the JMD benchmark. Additionally, some sounds fell well below the pure tone hearing thresholds, raising concern that the therapeutic levels may be inaudible, and thus too low to provide the intended benefit. These ranges indicate that variability in delivered therapeutic sound levels across brands and coupling conditions is potentially large enough to influence whether patients perceive sufficient benefit to initiate or continue sound therapy.

Systematic differences were observed across sound therapy levels. Although the sample size was too small to make definitive conclusions regarding the source of this variability, several factors may have contributed to the overall differences in output levels. The spectral shaping of the sound therapy may have influenced the overall energetic content of the signal. In this study, audiogram- or hearing loss-based shaping was selected when available. For manufacturers offering this feature, it is possible that manufacturer-specific software settings related to RECD implementation influenced the shaping of the sound therapy and contributed to the variability observed between brands. Depending on whether the fitting software was based on KEMAR-derived RECD values or average RECD values from older adult populations, identical sound therapy and insertion gain settings could produce different acoustic outputs. The effect of implementing individually measured CARL on-ear RECD values within the manufacturer's software was not investigated in the present study and may represent an important consideration for future research. Another factor which can influence hearing aid output is the potential for interactions with gain corrections for use of binaural hearing aids. Binaural corrections are adjustments applied to prescriptive targets to compensate for delivery of amplified sound to both ears simultaneously. The DSL v5 prescription applies a −3 dB correction regardless of hearing loss configuration; this was applied to the asymmetrical loss that was evaluated in this paper. Currently, it is unknown whether manufacturers provide or apply a binaural correction to the delivery of on-board sound therapy in hearing aids. The programming software used in this study did not offer an option to select such a feature.

In addition to broadband noise sound therapy signals evaluated in the present study, manufacturers also provide alternative sound therapy stimuli, including notch therapy (e.g., Signia), fractal or musical tones (e.g., Widex Zen Therapy), and nature-like sounds (e.g., Oticon, Resound). These sound therapies differ substantially in their spectral and temporal characteristics and may therefore interact differently with hearing aid amplification, coupling acoustics, and digital sound processing features. The present study focused specifically on steady-state broadband sound therapy approaches, which are available across different brands; however, future research should investigate whether similar acoustic and fitting considerations apply to other sound therapy paradigms.

4.2 Programming sound therapy and digital signal processing features

The present study demonstrates that sound therapy and noise reduction algorithms can interact counterproductively; this interaction may become more apparent as sound therapy levels and noise reduction strength increase. For at least one brand, the overall sound therapy intensity was reduced by 9 dB due to the NR algorithm, exceeding the 6 dB JMD benchmark. This indicates that NR has the potential to meaningfully influence patients' perceived benefit of their sound therapy. Power-sum averaging, which is more closely related to the overall perceived level, was more sensitive to the effects of noise reduction than arithmetic averaging. In contrast, arithmetic averaging provided greater sensitivity to localized spectral differences but may overemphasize changes in frequency regions that contribute less to perceived loudness. The findings indicate that the interaction between noise reduction and sound therapy is not consistent across devices and fitting conditions, warranting further consideration of the clinical implications of these differences.

Furthermore, current tinnitus management protocols suggest that sound therapy intensity should be selected by the user, with perceptual targets such as a “comfortable background level”, or more fixed points such as 6 dB sensation level (Henry et al., 2008; Jastreboff and Hazell, 2004; Jastreboff and Jastreboff, 2024). Sound therapy level is integral to tinnitus management, regardless of the therapeutic goal. The interaction between NR algorithms and sound therapy has the potential to alter the levels of the delivered therapy relative to clinicians' programming intentions and may interfere with user-directed adjustments, including those recommended in Tinnitus Retraining Therapy and Progressive Tinnitus Management (Henry et al., 2008; Jastreboff and Hazell, 2004). The interaction between noise reduction algorithms and sound therapy should be routinely assessed in clinical practice; to avoid this interaction altogether, it may be beneficial for clinicians to disable noise reduction in the sound therapy program. The latter strategy was automated in one brand that was evaluated in this study.

The maximum stable gain for the devices was minimally impacted with the presence of sound therapy, varying by only by 1–2 dB across the susceptible feedback frequencies (2,000–4,000 Hz). This suggests that, once the prescribed amplification is taken into account, the hearing aid system maintains a relatively stable functional feedback limit under each fitting configuration. The differences observed in MSG_ST values between the N4 and S2 conditions are likely not due to changes in the intrinsic maximum stable gain of the hearing aid system itself, rather, these differences reflect the influence of the programmed fitting parameters associated with each audiogram configuration. In the present study, measurements were conducted using the prescribed amplification, frequency shaping, and sound therapy settings specific to the N4 and S2 fittings. As a result, the gain distribution across frequencies differed between conditions, altering the amount of additional gain available before acoustic feedback occurred. The addition of sound therapy may have further contributed to these effects by increasing overall acoustic energy at frequencies susceptible to instability (2–4 kHz). Therefore, the observed variability in MSG_ST likely reflects the interaction between prescribed amplification, spectral shaping, and sound therapy characteristics, rather than changes in the underlying acoustic stability limit of the device itself.

The AGBF measure revealed a significant gain penalty associated with amplification to maintain gain stability and avoid acoustic feedback in the high-frequency regions (4,000–8,000 Hz). Across all conditions, the addition of sound therapy caused acoustic feedback to occur when amplification gain was set to prescriptive targets. This indicates that during programming, there will be a compromise among providing adequate levels of sound therapy, allowing audible levels of sound amplification, and avoiding audible feedback. In this regard, a secondary program specifically for tinnitus sound therapy is recommended in order to reduce the gain in high-frequency regions associated with feedback. Additionally, feedback reduction programs should be set to their maximum capability in the sound therapy program to minimize audible feedback.

4.3 Verification of sound therapy settings

Verification is an electroacoustic analysis of the hearing aid output; typically performed in the ear of the hearing aid wearer to assess the adequacy of the hearing aid output at the individual level. The Audiology Practice Standards Organization (APSO., 2021) states that a key component of fitting hearing aids for adult patients is verifying the hearing aid's output in the ear canal in accordance with frequency-specific prescriptive targets. Hearing aid verification yields advantages such as enhanced self-reported listening ability and improved speech intelligibility in both quiet and noisy environments for users (Almufarrij et al., 2021). In tinnitus research, verified hearing aid amplification outperformed unverified amplification in diminishing both the loudness and distress associated with tinnitus (Waechter and Jönsson, 2022). Although prescriptive targets for tinnitus sound therapy have not been developed, significant variability in hearing aid output compared to software programming has been demonstrated in a study using test-box measures (Hong et al., 2020) and on-ear measures in the present study. Brand-specific discrepancies between programmed and measured outputs were identified, with brands either surpassing or falling short of detection thresholds. The present electroacoustic characterization results also support the conclusion that default sound therapy levels are inconsistent, unpredictable, and vary with multiple factors. In clinical settings, this may lead to time-consuming troubleshooting and practice variability, thus underscoring the need for feasible verification methods for sound therapy. Hearing aid analyzers that offer real ear measurement of externally-generated signals can capture sound therapy intensity, frequency response, and the impact of DSP features (Hong et al., 2020; Rocha and Mondelli, 2020). This method of verification is clinically feasible, enabling clinicians to characterize the level of therapeutic sounds and can provide quantifiable parameters to document and utilize at later appointments.

4.4 Adapting APSO standards to sound therapy

Knowledge translation and clinical practice guidelines in audiology are imperative for delivering evidence-based practice. In the past, it has been noted that despite growing interest and efforts to produce high-quality practice guidelines, the lack of coordination among multiple practice organizations can be a barrier to the adoption of new practices (Moodie et al., 2011). Therefore, it is recommended that existing guidelines be adapted to serve as templates or starting points for guideline development and to reduce redundancy. Considering this, the following adaptations to APSO standards may extend the current guidelines to include programming of hearing aids that offer sound therapy (Table 3). The adaptations aim to support a systematic and patient-centered approach to fitting, programming, and verification of sound therapy in clinical practice (APSO., 2021). It should be noted that the considerations below pertain to fitting a hearing-aid-based sound therapy program, in addition to fitting the primary amplification and environment-specific programs provided by modern hearing aids.

5 Limitations

A manikin was used for on-ear measures in this study, but the manufacturer-specific programming software is designed for use with human ears. Evaluation on human ears may therefore serve to extend the generalizability of these findings. The nature of this study was exploratory; no repeat measures were conducted, which potentially impacted the replicability of results and generalizability of findings. However, the measurement procedures and hearing aid features used in this study have previously been independently evaluated and shown to demonstrate acceptable reliability in prior research. Overall, real ear measures are found to be reliable measures of sound delivered in the ear canal, up to and beyond 8,000 Hz with appropriate insertion depth of probe tube (Vaisberg et al., 2016). Real ear measures were also found to be reliable with repeated measures, with less than 2dB variation for frequencies 500–5,000 Hz. This indicates that with correct insertion depth, the real ear measures conducted in the present study were not likely to differ and that the observed differences are due to variation in hearing aid response rather than measurement error. Specifically, evaluation of NR strength was found to be within approximately 2 dB across repeated measurements, provided that the NR algorithm has sufficient time to reach its asymptotic performance (Scollie et al., 2016). In the present study, the NR algorithm was enabled for a period before recording to ensure compliance with the NR activation. Based on this, it is reasonable to assume that similar outcomes would be expected under a retest condition, although this was not directly verified within the current experimental design. Measures of feedback, MSG and AGBF, have also been previously evaluated for reliability. The procedures used in the present study closely follow those described by Ricketts et al. (2008) and adopted in Marcrum et al. (2018). It was found that feedback detection for MSG and AGBF measures was consistent within approximately 1 dB across repeated measurements using the same instrument in normal hearing listeners (Ricketts et al., 2008). This demonstrates the stability of these measures under repeated testing conditions and that the measures used in the present study demonstrate reasonable measurement reliability, supporting the assumption that observed differences are attributable to experimental conditions rather than measurement variability.

Another limitation of this study is that only an asymmetrical hearing loss configuration was evaluated, with the N4 audiogram simulated in the left ear and the S2 audiogram simulated in the right ear. Although none of the manufacturer's fitting software provided explicit options for binaural correction of sound therapy, it remains possible that proprietary binaural processing or device-to-device communication influenced the delivery of sound therapy in ways that were not apparent to the researchers and were therefore not assessed in this study. Consequently, the observed outputs may reflect interactions specific to the asymmetrical fitting configuration; the findings may not be directly generalizable to listeners with symmetrical hearing losses. Furthermore, the default sound therapy levels demonstrated limited audibility across frequencies, with most brands producing levels below hearing loss thresholds. It may be that these sounds are audible, given that the sound therapy was broadband and the hearing thresholds were based on narrowband signals. However, because a manikin was used for data collection, subjective ratings, such as the audibility, loudness, or sensation level of the sound therapy, were not captured. Further assessments on human subjects are recommended. Finally, the small sample size prevented inferential data analysis, limiting the presented results to descriptive summaries for two standard audiograms. Further study in larger samples would clarify the generalizability of these results.

6 Conclusions

Sound therapy delivered via a hearing aid for tinnitus management is affected by the hearing aid's physical coupling and signal-processing settings during fitting. As a result, hearing aid verification procedures are useful for characterizing sound therapy within the broader context of hearing aid fitting; this study has revealed that sound therapy levels vary across brands, hearing loss configurations, and coupling types. In some cases, these default settings may also interact counterproductively with core digital processing features. This paper outlines key considerations for fitting, programming, and verification of sound therapy with the aim of supporting more consistent clinical practice and improving patient access to effective care.

References

• 1

  AgnewJ. (1996). Acoustic feedback and other audible artifacts in hearing aids. Trends Amplif.1, 45–82. doi: 10.1177/108471389600100202

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 2

  AHead Simulations. (2024). CARL Basic. Available online at: https://aheadsimulations.com/collections/carl-units/products/carl-basic (Accessed May 31, 2026).

  • Google Scholar
  • View reference in article

• 3

  AlmufarrijI.DillonH.MunroK. J. (2021). Does probe-tube verification of real-ear hearing aid amplification characteristics improve outcomes in adults? A systematic review and meta-analysis. Trends Hear.25:2331216521999563. doi: 10.1177/2331216521999563

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 4

  American Speech-Language-Hearing Association. (1990). Audiometric Symbols. American Speech-Language-Hearing Association.Available online at: https://www.asha.org/policy/gl1990-00006/?srsltid=AfmBOopTs92bH8xOo0uuR2nar_l41nqDZXVipqIy5KfGtbFiU70waC2Y (Accessed May 31, 2026).

  • Google Scholar
  • View reference in article

• 5

  APSO. (2021). Hearing Aid Fitting Standard for Adult and Geriatric Patients. Audiology Practice Standards Organization. Available online at: https://www.audiologystandards.org/standards/publications/APSO_S2.1_Adult-HAF_20210502.pdf (Accessed May 31, 2026).

  • Google Scholar
  • View reference in article

• 6

  Audioscan. (2023). Verifit 2 Advanced Verification (Version 2023.2.5; Issue Version 2023.2.5). Etymonic Design Incorporated. Available online at: https://docs.audioscan.com/userguides/vf2manual.pdf (Accessed May 31, 2026).

  • Google Scholar
  • View reference in article

• 7

  Audioscan. (2026). Verifit^®VF-2 User's Guide 2025.1. Available online at: https://docs.audioscan.com/userguides/vf2manual.pdf (Accessed May 31, 2026).

  • Google Scholar
  • View reference in article

• 8

  BagattoM.MoodieS.ScollieS.SeewaldR.MoodieS.PlumfordJ.et al. (2005). Clinical protocols for hearing instrument fitting in the desired sensation level method. Trends Amplif.9, 199–226. doi: 10.1177/108471380500900404

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 9

  BallingL. W.JensenN.CaporaliS.CubickJ.SwitalskiW. (2019, December 18). Challenges of instant-fit ear tips: what happens at the eardrum? The Hearing Review. Available online at: https://hearingreview.com/hearing-loss/patient-care/hearing-fittings/challenges-of-instant-fit-ear-tips-what-happens-at-the-eardrum (Accessed May 31, 2026).

  • Google Scholar
  • View reference in article

• 10

  BisgaardN.VlamingM. S.DahlquistM. (2010). Standard audiograms for the IEC 60118-15 measurement procedure. Trends Amplif.14, 113–120. doi: 10.1177/1084713810379609

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 11

  BlauM.SankowskyT.StirnemannA.OberdannerH.SchmittN. (2008). Acoustics of open fittings. J. Acoust. Soc. Am. 123, 3011–3011. doi: 10.1121/1.2932603

  • CrossRef
  • Google Scholar

• 12

  BoisvertI.ClemeshaJ.LundmarkE.CromeE.BarrC.McMahonC. M. (2017). Decision-making in audiology: balancing evidence-based practice and patient-centered care. Trends Hear.21:2331216517706397. doi: 10.1177/2331216517706397

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 13

  British Society of Audiology. (2020). Fitting of Combination Hearing Aids for Subjects with Tinnitus. Available online at: https://www.thebsa.org.uk/wp-content/uploads/2020/10/Fitting-of-combination-hearing-aids-for-subjects-with-tinnitus.pdf (Accessed May 31, 2026).

  • Google Scholar
  • View reference in article

• 14

  ChungD.GannonP.MasonK. (1984). Factors affecting the prevalence of tinnitus. Audiology23, 441–452. doi: 10.3109/00206098409070084

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 15

  CimaR. F. F.MazurekB.HaiderH.KikidisD.LapiraA.NoreñaA.HoareD. J. (2019). A multidisciplinary European guideline for tinnitus: diagnostics, assessment, and treatment. HNO67, 10–42. doi: 10.1007/s00106-019-0633-7

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 16

  DaoA.FolkeardP.BakerS.PumfordJ.ScollieS. (2021). Fit-to-targets and aided speech intelligibility index values for hearing aids fitted to the DSL v5-adult prescription. J. Am. Acad. Audiol.32, 090–098. doi: 10.1055/s-0040-1718707

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 17

  dos SantosG. M.BentoR. F.de MedeirosI. R. T.OiticcicaJ.da SilvaE. C.PenteadoS. (2014). The influence of sound generator associated with conventional amplification for tinnitus control: randomized blind clinical trial. Trends Hear. 18:2331216514542657. doi: 10.1055/s-0034-1388704

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 18

  FolkeardP.AlamM.KochR.AbbasalipourP.ScollieS. (2023). An evaluation of the CARL manikin for use in “patient-free” real ear measurement: consistency and comparison to normative data. Int. J. Audiol.62, 376–382. doi: 10.1080/14992027.2022.2048104

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 19

  HenryJ. A. (2022). Sound therapy to reduce auditory gain for hyperacusis and tinnitus. Am. J. Audiol.31, 1067–1077. doi: 10.1044/2022_AJA-22-00127

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 20

  HenryJ. A.CarlsonK. F.TheodoroffS.FolmerR. L. (2022). Reevaluating the use of sound therapy for tinnitus management: perspectives on relevant systematic reviews. J. Speech Lang. Hear. Res.65, 2327–2342. doi: 10.1044/2022_JSLHR-21-00668

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 21

  HenryJ. A.SonstroemA.SmithB.GrushL. (2021). Survey of audiology graduate programs: training students in tinnitus management. Am. J. Audiol.30, 22–27. doi: 10.1044/2020_AJA-20-00140

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 22

  HenryJ. A.ZauggT. L.MyersP. J.SchechterM. A. (2008). Using therapeutic sound with progressive audiologic tinnitus management. Trends Amplif.12, 188–209. doi: 10.1177/1084713808321184

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 23

  HongT.RalliM.StockingC.SheppardA. (2020). Acoustic analysis of hearing aid sound therapy programs. Hear. Bal. Commun.18, 8–15. doi: 10.1080/21695717.2019.1692592

  • CrossRef
  • Google Scholar
  • View reference in article

• 24

  JastreboffP. J.HazellJ. W. P. (2004). Tinnitus Retraining Therapy: Implementing the Neurophysiological Model, 1st Edn. Cambridge, MA: Cambridge University Press. doi: 10.1017/CBO9780511544989

  • CrossRef
  • Google Scholar
  • View reference in article

• 25

  JastreboffP. J.JastreboffM. M. (2000). Tinnitus retraining therapy (TRT) as a method for treatment of tinnitus and hyperacusis patients. J. Am. Acad. Audiol.11, 162–177. doi: 10.1055/s-0042-1748042

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 26

  JastreboffP. J.JastreboffM. M. (2024). “Tinnitus retraining therapy,” in Textbook of Tinnitus, W. Schlee, B. Langguth, D. De Ridder, S. Vanneste, T. Kleinjung, and A. R. Møller, Eds (New York, NY: Springer International Publishing), pp. 589–616. doi: 10.1007/978-3-031-35647-6_47

  • CrossRef
  • Google Scholar
  • View reference in article

• 27

  KochkinS. (2007). MarkeTrak VII: obstacles to adult non-user adoption of hearing aids. Hear. J.60:24. doi: 10.1097/01.HJ.0000285745.08599.7f

  • CrossRef
  • Google Scholar
  • View reference in article

• 28

  KönigO.SchaetteR.KempterR.GrossM. (2006). Course of hearing loss and occurrence of tinnitus. Hear. Res.221, 59–64. doi: 10.1016/j.heares.2006.07.007

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 29

  KutybaJ.GosE.JedrzejczakW. W.Raj-KoziakD.KarpieszL.NiedziałekI.et al. (2022). Effectiveness of tinnitus therapy using a mobile application. Eur. Arch. Oto-Rhino-Laryngol.279, 1257–1267. doi: 10.1007/s00405-021-06767-9

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 30

  LangguthB.KreuzerP. M.KleinjungT.De RidderD. (2013). Tinnitus: causes and clinical management. Lancet Neurol.12, 920–930. doi: 10.1016/S1474-4422(13)70160-1

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 31

  LansbergenS.DreschlerW. A. (2020). Classification of hearing aids into feature profiles using hierarchical latent class analysis applied to a large dataset of hearing aids. Ear Hear.41, 1619–1634. doi: 10.1097/AUD.0000000000000410

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 32

  LelicD.CaporaliS.ParkerD.NielsenJ.BallingL. W. (2024). Impact of a combination sound therapy on tinnitus distress: an exploratory one-year longitudinal study. Front. Audiol. Otol. 2:2596. doi: 10.3389/fauot.2024.1322596

  • CrossRef
  • Google Scholar
  • View reference in article

• 33

  LiuH.ZhangJ.YangS.WangX.ZhangW.LiJ.YangT. (2021). Efficacy of sound therapy interventions for tinnitus management. Medicine.100:e27509. doi: 10.1097/MD.0000000000027509

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 34

  MarcrumS. C.PicouE. M.BohrC.SteffensT. (2018). Feedback reduction system influence on additional gain before feedback and maximum stable gain in open-fitted hearing aids. Int. J. Audiol.57, 737–745. doi: 10.1080/14992027.2018.1487088

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 35

  McNeillC.Távora-VieiraD.AlnafjanF.SearchfieldG. D.WelchD. (2012). Tinnitus pitch, masking, and the effectiveness of hearing aids for tinnitus therapy. Int. J. Audiol.51, 914–919. doi: 10.3109/14992027.2012.721934

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 36

  McSheffertyD.WhitmerW. M.AkeroydM. A. (2016). The just-meaningful difference in speech-to-noise ratio. Trends Hear. 20:2331216515626570. doi: 10.1177/2331216515626570

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 37

  MoodieS. T.KothariA.BagattoM. P.SeewaldR.MillerL. T.ScollieS. D. (2011). Knowledge translation in audiology: promoting the clinical application of best evidence. Trends Amplif.15, 5–22. doi: 10.1177/1084713811420740

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 38

  MyersP. J.GriestS.KaelinC.LegroM. W.SchmidtC. J.ZauggT. L.HenryJ. A. (2014). Development of a progressive audiologic tinnitus management program for veterans with tinnitus. J. Rehabil. Res. Dev.51, 609–622. doi: 10.1682/JRRD.2013.08.0189

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 39

  NewmanC. W.SandridgeS. A.MeitS. S.CherianN. (2008). Strategies for managing patients with tinnitus: a clinical pathway model. Semin. Hear.29, 300–309. doi: 10.1055/s-0028-1082035

  • CrossRef
  • Google Scholar
  • View reference in article

• 40

  RickettsT.JohnsonE.FedermanJ. (2008). Individual differences within and across feedback suppression hearing aids. J. Am. Acad. Audiol.19, 748–757. doi: 10.3766/jaaa.19.10.3

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 41

  RickettsT. A.BentlerR.MuellerH. G. (2017). Essentials of Modern Hearing Aids: Selection, Fitting, and Verification. San Diego, CA: Plural Publishing.

  • Google Scholar

• 42

  RochaA. V.MondelliM. F. C. G. (2020). Applicability of the real ear measurement for audiological intervention of tinnitus^,. Brazil. J. Otorhinolaryngol.86, 14–22. doi: 10.1016/j.bjorl.2018.07.010

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 43

  SchleeW.LangguthB.De RidderD.VannesteS.KleinjungT.MøllerA. R. (Eds.). (2024). Textbook of Tinnitus. New York, NY: Springer International Publishing. doi: 10.1007/978-3-031-35647-6

  • CrossRef
  • Google Scholar
  • View reference in article

• 44

  ScollieS.LevyC.PourmandN.AbbasalipourP.BagattoM.RichertF.et al. (2016). Fitting noise management signal processing applying the American Academy of Audiology pediatric amplification guideline: verification protocols. J. Am. Acad. Audiol.27, 237–251. doi: 10.3766/jaaa.15060

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 45

  ScollieS.SeewaldR.CornelisseL.MoodieS.BagattoM.LaurnagarayD.et al. (2005). The desired sensation level multistage input/output algorithm. Trends Amplif.9, 159–197. doi: 10.1177/108471380500900403

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 46

  ShekhawatG. S.SearchfieldG. D.KobayashiK.StinearC. M. (2013). Prescription of hearing-aid output for tinnitus relief. Int. J. Audiol.52, 617–625. doi: 10.3109/14992027.2013.799787

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 47

  ShettyH. N.BasavarajV. (2022). Amplification strategies to reduce tinnitus: a paired comparison method. J. Int. Adv. Otol. 18, 8–13. doi: 10.5152/iao.2022.20057

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 48

  TylerR. S.PerreauA.PowersT.WattsA.OwenR.JiH.ManciniP. C. (2020). Tinnitus sound therapy trial shows effectiveness for those with tinnitus. J. Am. Acad. Audiol.31, 6–16. doi: 10.3766/jaaa.18027

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 49

  VaisbergJ. M.MacphersonE. A.ScollieS. D. (2016). Extended bandwidth real-ear measurement accuracy and repeatability to 10 kHz. Int. J. Audiol.55, 580–586. doi: 10.1080/14992027.2016.1197427

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 50

  WaechterS.JönssonA. (2022). Hearing aids mitigate tinnitus, but does it matter if the patient receives amplification in accordance with their hearing impairment or not? a meta-analysis. Am. J. Audiol.31, 789–818. doi: 10.1044/2022_AJA-22-00004

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article