Abstract
Purpose:
We present results from a 6-month field trial of a transitional intervention for debilitating primary hyperacusis, including a combination of structured counseling; promotion of safe, comfortable, and healthy sound exposure; and therapeutic broadband sound from sound generators. This intervention is designed to overcome barriers to successful delivery of therapeutic sound as a tool to downregulate neural hyperactivity in the central auditory pathways (i.e., the maladaptive mechanism believed to account for primary hyperacusis) and, together with the counseling, reduce the associated negative emotional and physiological reactions to debilitating hyperacusis.
Method:
Twelve adults with normal or near-normal audiometric thresholds, complaints consistent with their pretreatment loudness discomfort levels ≤ 75 dB HL at multiple frequencies, and hearing questionnaire scores ≥ 24 completed the sound therapy–based intervention. The low-level broadband therapeutic sound was delivered by ear-level devices fitted bilaterally with either occluding earpieces and output-limiting loudness suppression (LS; to limit exposure to offensive sound levels) or open domes to maximize comfort and exposure to sound therapy. Thresholds for LS (primary outcome) were incrementally adjusted across six monthly visits based on treatment-driven change in loudness judgments for running speech in sound field. Secondary outcomes included categorical loudness judgments, speech understanding, and questionnaires to assess the hyperacusis problem, quality of life, and depression. An exit survey assessed satisfaction with and benefit from the intervention and the counseling, therapeutic sound, and LS components.
Results:
The mean change in LS (34.8 dB) was highly significant (effect size = 2.045). Eleven of 12 participants achieved ≥ 16-dB change in LS, consistent with highly significant change in sound-based questionnaire scores. Exit surveys indicated satisfaction with and benefit from the intervention.
Conclusion:
The transitional intervention was successful in improving the hyperacusis conditions of 11 of 12 study participants while reducing their sound avoidance behaviors and reliance on sound protection.
This report is the final in a series of four companion reports in which we describe the results from a 6-month field trial of a patented transitional intervention for debilitating loudness-based hyperacusis (LH; Eddins et al., 2020). Collectively, these reports provide: the background and rationale for the transitional sound-based intervention (Formby et al., 2024); a detailed description of the structured counseling protocol (Cherri et al., 2024); and an overview of the treatment device, the fitting protocol, and our strategy for delivery of therapeutic sound in low- and high-risk sound environments (Eddins et al., 2024). Together, these counseling and sound-based device protocols were the primary components of our transitional intervention. We present compelling evidence in this report that these protocols can be successfully combined to overcome barriers to the delivery of therapeutic sound, which is our primary tool to recalibrate (i.e., downregulate) LH-inducing hyper-gain neural activity within the central auditory pathways. The counseling protocol, along with the progressive protective sound management approach and the therapeutic sound component of the intervention, will be shown here to contribute to statistically and clinically significant reduction of LH and the associated negative reactions. We also highlight an adaptive treatment-driven strategy for adjustment of output-limiting sound protection over the course of the trial. We quantify the magnitude of this adaptive adjustment and present it as our primary outcome measure along with a battery of secondary outcome indices. These primary and secondary outcome measures consistently reveal positive treatment-related change in sound tolerance and clinically meaningful improvements in the conditions of 11 of the 12 hyperacusis patients who participated in this trial, including individuals with pretreatment loudness discomfort levels (LDLs) as low as 35 dB HL.
Study Hypothesis
The hypothesis evaluated in this proof-of-concept trial is that individuals with primary debilitating LH can be successfully treated using the transitional intervention described above and implemented and evaluated in the study protocol detailed below. Successful treatment will be evidenced by quantifiable improvements in sound tolerance and decreased reliance on sound protection. In addition, we hypothesize that, posttreatment, participants will demonstrate greater willingness to accept healthy sound and be exposed to sound environments that previously were avoided. At the conclusion of the intervention, treatment-related subjective improvements and benefit also will be revealed in positive responses to self-report questionnaires. Likewise, we hypothesize favorable feedback from an exit survey designed to assess satisfaction with the intervention and benefits from each of the primary contributing components to the intervention. Furthermore, we anticipate that treatment-related improvement in sound tolerance will offer opportunities for enhanced speech perception posttreatment.
Method
Overview of the Field Trial
The field trial of our transitional intervention began in September of 2019 and ended in December of 2021. This trial period coincided with the primary period of the COVID-19 outbreak in the United States and worldwide, requiring pauses in the study and modifications of the originally planned trial protocol and procedures as described below. The study implementation and evaluation protocol (i.e., the study protocol) for the field trial is summarized in Table 1 with complete details later in this report.
Table 1.
Study protocol: Summary of study visits, timeline, and procedures.
| Visit | Timeline | Procedures completed |
|---|---|---|
| Intake |
|
|
| Fitting and counseling (T0) | 2 weeks after intake |
Day 1: Device programming
|
| Follow-up (T1–T6) | Monthly |
|
Note. MoCA = Montreal Cognitive Assessment; SRT = Speech Reception Threshold; LDL = loudness discomfort level; Cat4 and Cat6 = Contour Categorical Loudness Judgments for “comfortable” and for “loud, but OK”, respectively; NU-6 = Northwestern University Auditory Test No. 6; HQ = Hyperacusis Questionnaire; NAQ = Noise Avoidance Questionnaire; GÜF = English version of the German Index of Sound-Related Distress; TS-H = Tampa Scale for Hyperacusis; QoLI = Quality of Life Inventory; BDI-II = Beck Depression Inventory–II; SG = sound generator; LS = loudness suppression; HINT = Hearing in Noise Test.
Participants
This National Institutes of Health (NIH)–funded proof-of-concept trial was designed with a target of completing data collection for 10 participants. Eighteen prospective candidates were recruited and screened for debilitating LH per the specified inclusion/exclusion criteria listed in Table 2. Recruitment included postings on the Hyperacusis Research Limited website, hyperacusis-relevant Facebook pages, and Craigslist. In addition, communication with local referral sources, professional colleagues, and the hyperacusis community were fruitful for identifying some of the prospective candidates for the trial. Two of these 18 candidates failed to meet eligibility criteria, two others experienced reactive tinnitus and withdrew after the device fitting visit (described below), one participant relocated out of state and withdrew from the trial, and another postponed participation due to unrelated personal conflicts. All candidates provided approved informed consent per institutional review board guidelines at the University of South Florida, where all laboratory procedures and testing were conducted. All participants were negative for significant cognitive impairment on the Montreal Cognitive Assessment (Nasreddine et al., 2005).
Table 2.
Study inclusion and exclusion criteria.
| Inclusion criteria | Exclusion criteria |
|---|---|
|
|
Note. LDL = loudness discomfort level; dB HL = decibel hearing level; BBN = broadband noise; MoCA = Montreal Cognitive Assessment.
At intake, all candidates reported complaints consistent with abnormally reduced sound tolerance, routine use of hearing protection devices (HPDs), self-isolation, or both, to avoid offending sounds in their typical daily environments. They also related varying subjective complaints of sound-induced aural pain, physical discomfort, distress, or some combination of these, confirmed by their responses on self-report questionnaires described below. The primary screening criteria for LH were: (a) LDLs ≤ 75 dB HL at two or more audiometric frequencies in at least one ear and (b) a corresponding score ≥ 24 on the Hyperacusis Questionnaire (HQ; Khalfa et al., 2002). The LDL criterion is consistent with values commonly cited as evidence of primary hyperacusis, including reported LDLs of 76.9 (Anari et al., 1999), 79.8 (Nelting et al., 2002), and 71.9–73.5 dB HL (Jüris et al., 2014) across a range of studies. Sherlock and Formby's (2005) analysis of normative LDLs also supports this criterion. They reported that the lower end of the 95% confidence interval for typical LDLs was ≥ 75 dB HL for audiometric frequencies between 500 and 4000 Hz. Our LDL criterion also agrees well with that for hyperacusis proposed by Aazh and Moore (2017). They reported that an HQ cutoff score of 22 coincided with the lower end of the 95% confidence interval for which their patients had both (abnormally) reduced LDLs (≤ 77 dB HL) and associated hyperacusis complaints. Khalfa et al. (2002), who developed the HQ, proposed a more stringent cutoff score of 28 for identification of severe hyperacusis. However, the latter HQ cutoff score has been challenged for being too strict (Aazh & Moore, 2017; Fackrell et al., 2015; Fioretti et al., 2015), leading us to select a compromise HQ cutoff score ≥ 24 as the criterion for participant inclusion in our field trial. (We waived the HQ criterion for one participant, who during the peak COVID-19 shutdown period was undergoing unrelated medical problems and, while homebound, scored 20 on the HQ. Otherwise, his hyperacusis profile was consistent with those of the other participants who qualified for the trial.) Thus, our criteria for debilitating LH and inclusion in this trial were: subjective complaints consistent with abnormally reduced sound tolerance (including routine use of HPDs, self-isolation from typical sounds, or both); associated distress, untoward sound-induced discomfort (including pain), or both; LDLs ≤ 75 dB HL at multiple frequencies in at least one ear; and an HQ score ≥ 24.
Ultimately, 12 participants (eight men and four women, ages 22–67 years) completed either most or all scheduled study activities described below. Their demographic information and histories revealed a range of medical and audiological conditions, which are briefly summarized for the group in Table 3. Among the participants, traumatic head injury and significant noise exposure were the most reported causes of their hyperacusis conditions.
Table 3.
Participant demographic and medical and comorbidity summary.
| # | Age at enrollment (in years) |
Sex | Duration of hyperacusis (in years) |
Tinnitus | Medical and comorbid history |
|---|---|---|---|---|---|
| 1 | 24.7 | M | 1.1 | Yes |
|
| 2 | 60.9 | F | 7.0 | No |
|
| 3 | 28.5 | F | 16.5 | Yes |
|
| 4 | 67.0 | M | 20+ | No |
|
| 5 | 61.4 | M | 5.5 | No |
|
| 6 | 49.7 | F | 3.5 | No |
|
| 7 | 22.5 | M | 2.0 | Yes |
|
| 8 | 27.7 | M | 5.8 | Yes |
|
| 9 | 40.2 | M | 21.0 | Yes |
|
| 10 | 33.4 | M | NK | No |
|
| 11 | 32.3 | M | 1.9 | Yes |
|
| 12 | 28.3 | F | 2.0 | No |
|
Note. # = participant number; M = male; TMJD = temporomandibular joint disorder; F = female; POTS = Postural Orthostatic Tachycardia Syndrome; EDS = Ehlers–Danlos Syndrome; MVA = motor vehicle accident; PTSD = posttraumatic stress disorder; NK = not known; ASD = autism spectrum disorder.
Shown in Figure 1 are the mean and range of the baseline audiometric air-conduction thresholds measured for the 12 participants. Their individual thresholds and LDLs are provided in Appendix A. All participants had audiometric thresholds within the prescribed range for eligibility (no more than a mild-to-moderate hearing loss through 4000 Hz). Eleven of the 12 participants had audiometric thresholds ≤ 30 dB HL, bilaterally, across frequency. Qualifying LDLs among the participants, measured per the ascending-level protocol described by Sherlock and Formby (2005), ranged from a low value of 30 dB HL up to the criterion cutoff value of 75 dB HL. The ear with the lowest LDLs was designated the test ear (TE) for the trial. An additional audiometric inclusion criterion (see Table 2), met by all qualifying participants, was a normal (negative) decay response for low-level (10 dB SL) broadband noise presented continuously for 1 min. This modified “tone decay” measurement (Carhart, 1957) was included as an additional eligibility criterion to rule out abnormal loudness adaptation to “soft, but comfortable” noise of the kind used in our therapeutic sound protocol.
Figure 1.
Audiometric thresholds: group mean pure-tone air-conduction thresholds for the 12 participants' right (red symbols) and left (blue symbols) ears measured at the intake visit. The ranges of the lowest and highest individual thresholds are shown at each frequency. L = left; R = right.
Interventions
Three basic interventional elements contributed to the transitional protocol in this study.
Structured counseling, delivered in the scripted format detailed by Cherri et al. (2024), served to educate the participant about hearing, sound, loudness, hearing loss, and LH. The counseling was designed to prepare the participant to transition from the counterproductive overuse of HPDs and corresponding self-imposed sound isolation to the starting point for treatment marked by the initiation of protective sound management and the delivery of enriched controlled sound therapy. The counselor explained the rationale for the use of therapeutic sound and the strategy underlying our transitional treatment approach, which promoted the acceptance of enriched sound (and the avoidance of silence) to induce recalibration of hyper-gain auditory processes that are now believed to give rise to LH. The counseling also was an essential component for addressing and promoting treatment-related reduction in the negative emotional and physiological reactions associated with LH-related distress and stress responses.
Protective sound management, described in detail by Eddins et al. (2024), was achieved with a pair of custom sound-attenuating, body heat–activated, expanding, stented ear plugs (Juneau et al., 2010) designed to protect against offending sound levels. The attenuating effects of the custom ear plugs were, in turn, offset by unity-gain amplification achieved with digitally programmable behind-the-ear (BTE) instruments to provide “transparent” and comfortable exposure to low and moderate level sounds. Participant exposure to potentially offending high-level sounds was restricted by output-limiting LS, which was achieved with programmable adjustment of the activation threshold for the LS set in the digital instruments. The output-limiting setting and its release were based on treatment-related incremental changes in participant loudness judgments for “loud, but OK” running speech over the course of the intervention. These loudness judgments were performed in sound field with the participant using an active device in one ear while the other ear was occluded. Thus, “LS release” reflects an increase in the activation threshold of the output-limiting settings implemented in the protective devices following treatment-driven improvements in the participant's loudness tolerance. Accordingly, these treatment-driven improvements directly reflect the participant's loudness judgments for running speech in sound field.
Therapeutic sound, from sound generators built into the BTE instruments, was delivered as continuous low-level neutral broadband white noise at levels judged to be “soft, but comfortable” bilaterally. The therapeutic sound from the sound generators was supplemented by healthy enriched environmental sound from neutral low-level sound sources in the home (e.g., fans, personal sound generators, etc.). Two options were available to the participants for the use of the device sound generators. In controlled sound environments, which placed the participants at low risk for offending high-level exposures, the sound generators delivered the therapeutic sound via a thin-tube, open-dome, earmold configuration, whereas, in high-risk, uncontrolled sound environments, each participant had the option for use of the therapeutic sound in combination with protective sound management using the thin tube coupled to the custom closed ear plug. The fitting and implementation of the protective sound management and treatment devices, described in detail by Eddins et al. (2024), were engineered for this study by General Hearing Instruments, Inc., recently reincorporated as Soft Touch Labs.
Study Visits and Activities
Intake Visit
The study protocol began with an initial assessment of each candidate at an intake visit. These assessment activities are listed in Table 1. The primary purpose of this visit was to obtain consent, determine and confirm prospective participant interest and eligibility for the trial through interview, review relevant history, and complete study questionnaires. Additionally, study candidates completed the HQ and a traditional audiometric test battery, including measures of air- and bone-conduction thresholds, speech reception thresholds, word recognition, tympanometry, and a modified tone decay test with broadband noise. A set of loudness measurements for stimuli delivered through supra-aural headphones (model HDA-200, Sennheiser) also was completed for each ear, including the primary screening measure, LDLs, as well as Contour Test loudness judgments (Cox et al., 1997). These latter categorical loudness judgments at the intake visit and at subsequent follow-up visits (described below) were measured with pulsed warble tones and spondee words presented in three ascending tracks, each using incremental 5-dB steps beginning with judgments of “very soft” per Cox et al. (1997). We limited all stimulus levels in each ascending track to threshold judgments of “loud, but OK” to avoid potentially offending sound levels. For the spondees and warble tones presented at 500 and 2000 Hz, Contour Test measurements focused on judgments of “comfortable” and the threshold for the “loud, but OK” category. The full Contour Test protocol was measured for a 4000-Hz warble tone, with termination at the threshold judgment for “loud, but OK.” These loudness measurements, administered by manual adjustment of the stimulus presentation using an audiometer, served to establish the TE for our primary analyses, quantify the severity and extent of the candidate's sound tolerance problem, confirm our primary eligibility screening measures, and familiarize the prospective participants with our secondary outcome measures. Lastly, for eligible participants, earmold impressions were completed bilaterally and shipped for production of custom ear plugs (see Eddins et al., 2024, for a detailed description of the impression process).
Counseling and Device Fitting Visit (T0)
Approximately 2 weeks after completion of the intake visit, each qualifying participant returned for a 2-day study visit during which scheduled baseline outcome measures were collected (see Table 1). On Day 1 of the visit, prescribed auditory measures, including loudness judgments and word recognition test measures, were completed in sound field without the treatment devices. Subsequently, the device fitting protocol, described in detail by Eddins et al. (2024), was executed. The fitting protocol included the initial setting and activation of both the output-limiting protective LS and the therapeutic sound generators.
Briefly, the LS setting component of the fitting protocol was performed with running speech delivered in sound field at a high level (usually at 80 dB SPL to simulate a loud “real world” signal). The output of the active protective treatment device, while in use by the participant who was fitted with the occluding custom ear piece, was initially reduced to the minimum output setting. The device in the other ear was inactive and served as an occluding earplug during these sound field measurements. With the running speech signal (“Carrot Passage”) continuously playing in a loop, the LS setting was systematically released in 1- or 2-dB steps (i.e., the active device output setting was increased as the output-limiting threshold was incrementally raised) until the participant initially judged the output level of the device as “loud, but OK” for the running speech. This protocol then was repeated to set the output-limiting LS in the opposite-ear device.
The sound generator setting component of the fitting protocol was performed by initially adjusting the sound generator to the minimum output setting. The sound generator level was increased in 1- or 2-dB steps until the participant indicated that the loudness was “soft, but comfortable.” This level was recorded, and the output setting was further increased until the participant indicated that the level was “comfortable.” The sound generator was set between these two levels, and the participant confirmed that the final level was acceptable to them. The opposite-ear device then was set in the same way, and the sound generator output levels were then adjusted as necessary to achieve a judgment of equal loudness between the ears with both devices activated.
After completing the device fitting protocol, scheduled loudness judgments for the warble tone and spondee stimuli (Hirsh et al., 1952) and outcome measures for the Hearing in Noise Test (HINT; Nilsson et al., 1994) and the Northwestern University Auditory Test No. 6 (NU-6) word recognition test (Tillman & Carhart, 1966) were completed for each participant using the protective treatment devices in sound field. The participants then returned the study devices to the study audiologist at the end of this visit.
On Day 2 of the counseling and fitting visit, baseline study questionnaires (see Table 1) were completed. The study audiologist provided the structured counseling component of the transitional intervention as detailed by Cherri et al. (2024). The counseling session was followed by a period of instruction during which the study audiologist practiced with the participant device insertion and exchange of the thin tubing between the open-dome and closed ear plug configurations; demonstrated muting of the LS protective operation (and sound generator output) by a toggle control on the device (rendering the custom ear plug a simple HPD); described device use (encouraging at least 8 hr of daily usage supplemented by exposure to enriched environmental sound), care, and cleaning of the open dome and ear plug configurations, including the thin-tube coupling; and requested the participant to prepare a log (diary) to document daily use of the two coupling configurations during the trial. Finally, participant questions were answered by the study audiologist, who previewed and scheduled the first of the six monthly follow-up visits (T1), after which participants were sent home with their study devices.
Follow-Up Treatment Visits (T1–T6)
Participants completed six follow-up treatment visits, designated T1–T6, each scheduled approximately 1 month apart, over the 6-month transitional intervention period. Scheduled study outcome measurements and data collection, treatment-driven device adjustments to release LS, reinforcement counseling, and other study-related activities were completed at each follow-up visit as shown in Table 1.
All follow-up treatment visits included scheduled loudness judgments for Contour Test Category 6 (threshold judgment for “loud, but OK”) pulsed warble tones measured at 500 and 2000 Hz and loudness judgments for the Contour Test Categories 1 (“very soft”) to 6 (threshold judgment for “loud, but OK”) for warble tones presented at 4000 Hz. The LS setting component of the fitting protocol was repeated for each ear as performed on Day 1 of the counseling and device fitting visit, and the new level was documented and programmed into the protective treatment devices at that follow-up visit.
Reinforcement counseling was provided as needed at each follow-up visit. The output levels from the sound generators were checked and reaffirmed with each participant to be soft and comfortable, in which case no adjustment was made (unless otherwise indicated, no adjustment was needed for most participants over the course of the trial). The study audiologist reviewed the participant's use of enriched sound sources at home and outside of the home, including the use of both coupling options (open dome or occluding earpieces) with the sound generators (as logged in the usage diary). The continued use of neutral, low-level output devices (that the participant reported to be comfortable and enjoyable to use in the home) was encouraged. Participants were encouraged to increase their exposure gradually to healthy sound-enriched conditions and to avoid quiet environments. At the end of each visit, participant questions and concerns were answered and addressed by the study audiologist. The participant was encouraged to share problems and challenging conditions encountered between study visits as well as situations for which the transitional intervention and treatment devices were proving beneficial or problematic.
Visits T3 and T6
In addition to the above study activities and measurements made at all follow-up visits, tasks completed at the T3 and T6 visits included the questionnaire battery, pure-tone air-conduction thresholds, loudness judgments for warble tones and spondees, and word recognition test measures in quiet and in noise, with and without the devices (see Table 1). Follow-up visit T6 was the final study visit and concluded with completion of the exit survey. Study devices were collected, and end-of-treatment counseling and relevant resource materials were provided for the participant. The latter materials included a summary of the original counseling session (see Cherri et al., 2024), literature relevant to living with hyperacusis, associated website listings, recommendations for sound enrichment options (devices, sound files, or both), and online resources for additional support.
Virtual Follow-Up Visits
Virtual visits were offered for four participants from April to June of 2020, in response to a COVID-19–related university-wide lockdown. These visits were offered remotely via Microsoft Teams while in-lab visits were postponed or when participants could not attend their in-person visits due to illness or exposure. The virtual visit protocol is shown in Table 4. The virtual visits allowed three participants to increase the output-limiting threshold settings in their devices at least once during the trial period without incident. There were no instances in which participants rejected any of their virtual visit programmed device changes.
Table 4.
Virtual visit criteria for adjustments to programming.
| Change to device | Criteria |
|---|---|
| Lower the LS level |
|
| LS level unchanged |
|
| Increase the LS level |
|
| SG level unchanged |
|
| SG level increased |
|
Note. LS = loudness suppression; SG = sound generator; TS-H = Tampa Scale for Hyperacusis.
Primary Outcome
The primary outcome was a change measure of the output-limiting activation threshold set in the protective treatment devices, quantified here as LS release (in dB). Note that the LS terminology follows from Sammeth et al. (2000), who coined the use of LS in their application of output limiting for hyperacusis, reflecting suppressed loudness perception for high-level inputs. This treatment-driven change measure reflects improvements in sound tolerance and follows directly from change in each participant's loudness judgments for “loud, but OK” running speech measured at each monthly follow-up visit T1–T6 over the course of the intervention. This loudness judgment for running speech, which offered a longer stimulus for setting the output-limiting threshold for LS, also provided us a loudness judgment for a stimulus with greater “real world” relevance than loudness judgments for brief tones or single utterances; the latter have traditionally been used in related clinical research, and these are included below as secondary outcomes. Thus, incremental changes in the treatment-driven loudness judgments for running speech from one follow-up visit to the next determined the incremental changes implemented in the output-limiting activation thresholds (i.e., the LS release) programmed into the protective treatment devices across the follow-up visits.
The total magnitude of LS release, representing the cumulative incremental change in the output-limiting activation threshold settings programmed into the TE protective treatment device between the baseline visit (T0) and final visit (typically the T6 follow-up visit), is reported here as our primary outcome measure. Although we documented LS release independently for each ear over the course of the intervention, unless indicated otherwise, we report here the primary outcome and secondary outcomes only for each participant's designated TE (i.e., in this trial, the participant's ear with the poorest pretreatment sound tolerance). This follows standard practice in clinical trials research to select one study measure, a priori, and to designate the results for that study measure, collected for a specified ear (i.e., the study TE), as the trial's primary study outcome (see Scherer & Formby, 2019). Other outcome measures collected for the designated TE and for the non-TE are considered secondary outcomes. Thus, per our study hypothesis, a clinically and statistically meaningful incremental change in the primary outcome between T0 and the final follow-up visit provides both objective and functional evidence of improved sound tolerance, reduced reliance on sound protection, greater acceptance of exposure to everyday sounds, and positive support for the efficacy of our transitional intervention for LH.
Secondary Outcomes
Audiometric and Auditory Measures
Scheduled audiometric and auditory measures were collected as shown in Table 1 to evaluate treatment-related changes in auditory function at the various study visits. These measures included: audiometric pure-tone thresholds; speech reception thresholds for CID W-22 spondee words (Hirsh et al., 1952); categorical loudness judgments for warble tones and spondees collected per Cox et al. (1997); and NU-6 word recognition (Tillman & Carhart, 1966) and HINT (Nilsson et al., 1994) measures. The NU-6 and HINT measures were presented at two levels, corresponding to the participant's judgments of Category 4 and threshold Category 6 levels for spondees. A priori, we expected little or no treatment-related change in the pure-tone thresholds, which were administered primarily to monitor change in each participant's audiometric status over the course of the trial. We hypothesized that treatment-related improvements in the speech understanding measures in quiet and noise would be achieved over the course of the intervention by the participants. This hypothesis, however, was based on participant performance not being restricted by their pretreatment performance at or near the ceiling of the measurement range for these speech tests.
Study Questionnaires
Six self-report questionnaires were completed/collected over the course of the study at the intake, T0, T3, and T6 visits to evaluate treatment-related subjective change. Four were sound-based questionnaires, and two were non–sound-based questionnaires. All six questionnaires were collected at each of these visits. Scores for the respective questionnaires, including cutoff values and ranges of scores for categories of severity (if available in the literature) as proposed by the developing investigators, are summarized in Table 5. 1 We hypothesized that a successful intervention would be accompanied by statistically improved change scores on each of these questionnaires between the participants' T0 and T6 visits.
Table 5.
Summary of study questionnaires with possible range of scores and corresponding categories of severity.
| Questionnaire | Origin | Purpose | Range | Categorization of severity | |||
|---|---|---|---|---|---|---|---|
| HQ | Khalfa et al. (2002) | Hypersensitivity to sounds | 0–42 | No hyperacusis | Presence of hyperacusis | ||
| 0–27 | 28–42 | ||||||
| GÜF | Bläsing et al. (2010); Nelting & Finlayson (2004) | Sound-related distress and annoyance | 0–45 | Mild | Moderate | Serious | Severe |
| 0–9 | 10–15 | 16–23 | 24–45 | ||||
| TS-H | Jüris et al. (2014) | Sound-related pain complaints | 17–68 | Low | High | ||
| 17–36 | 37–68 | ||||||
| NAQ | Blaesing & Kroener-Herwig (2012) | Sound avoidance behaviors | 0–100 | No rating criteria | |||
| QoLI | Frisch et al. (2005) | Overall quality of life | 0–77 | Very low | Low | Average | High |
| 0–37 | 38–43 | 44–58 | 59–77 | ||||
| BDI-II | Beck et al. (1996) | Depressive symptoms | 0–63 | Minimal | Mild | Moderate | Severe |
| 0–13 | 14–19 | 20–28 | 29–63 | ||||
Note. HQ = Hyperacusis Questionnaire; GÜF = English version of the German Index of Sound-Related Distress; TS-H = Tampa Scale for Hyperacusis; NAQ = Noise Avoidance Questionnaire; QoLI = Quality of Life Inventory; BDI-II = Beck Depression Inventory–II.
Sound-Based Questionnaires
The HQ is one of the earliest developed and one of the most widely reported subjective tools for assessing supra-threshold auditory hypersensitivity (Khalfa et al., 2002). The HQ has been used both as a screening tool for exclusion of participants with hyperacusis and as an outcome measure of treatment-related change in the severity of hyperacusis (see Fackrell et al., 2015). We applied the HQ in both ways in this study. The HQ is based on 14 negatively worded items, each rated on a 4-point response scale with scores of 0 (no), 1 (yes, a little), 2 (yes, quite a lot), and 3 (yes, a lot). The total scoring range spans from 0 to 42 with attentional, social, and emotional subscales. Khalfa et al. (2002) proposed an HQ global score ≥ 28 as a cutoff indicator of a significant hyperacusis condition, with greater scores indicative of a greater problem.
The Questionnaire on Hypersensitivity to Sound (GÜF) is a tool developed by Nelting and Finlayson (2004) and validated in tinnitus patients by Bläsing et al. (2010). In this study, we used the translation of the questionnaire from the original German to English as reported in Bläsing et al. (2010). The GÜF was designed to evaluate subjective distress/annoyance associated with hypersensitivity to sound using 16 items, each rated on a 4-point response scale with scores of 0 (never correct), 1 (sometimes correct), 2 (often correct), and 3 (always correct). The GÜF total score range spans from 0 to 45, with four levels of distress (mild to severe) and three subscales (cognitive reactions to hyperacusis, actional/somatic behavior, and emotional reaction to external noises).
The Tampa Scale of Kinesiophobia (TS-K; Miller et al., 1991) is a questionnaire originally developed as a 17-item index of excessive, irrational, and debilitating fear of movement, physical activity, and fear avoidance related to (re)injury in chronic pain patients with musculoskeletal disorders. Jüris et al. (2014) adapted the TS-K for use as a subjective tool to assess pain of suffering due to sound exposure. With assistance from Jüris (personal communication, 2019), we have translated their Swedish version for use with English-speaking hyperacusis patients. Here we refer to our English version (see Appendix B) as the Tampa Scale for Hyperacusis (TS-H; Formby & Eddins, 2019). The TS-H follows the 17-item format and scoring of the original TS-K , using a 4-point response scale with scores of 1 (strongly disagree), 2 (disagree), 3 (agree), and 4 (strongly agree). Higher scores indicate hyperacusis of greater severity; the cutoff for a clinically meaningful condition is ≥ 37 (following the cutoff criterion set for the TS-K).
The Noise Avoidance Questionnaire (NAQ) assesses the avoidance of sound in daily life and those events and places avoided (Blaesing & Kroener-Herwig, 2012; English translation). The first 15 of the 25 items address avoidance behaviors using the introductory phrase, “In order to expose myself to no or less noise . . . ”; the remaining 10 items, 16–25, address the events and places avoided using the introductory phrase, “In order to expose myself to no or less noise I avoid . . . ” Each item is ranked on a 5-point scale with scores of (0) never, (1) rarely, (2) occasionally/sometimes, (3) often, and (4) very often/always. The NAQ scores range from 0 to 100, with a higher score indicating a greater sound- or noise-related avoidance problem (Blaesing & Kroener-Herwig, 2012).
Non–Sound-Based Questionnaires
The Quality of Life Inventory (QoLI) is a 32-item tool previously used by Jüris et al. (2014) in hyperacusis research. The QoLI assesses importance and satisfaction for 16 areas of life (e.g., health, self-esteem, etc.; Frisch et al., 2005). Importance is ranked on a 3-point scale with scores of (0) not important, (1) important, or (2) extremely important. Satisfaction is ranked on a 6-point scale, including three possible responses for Dissatisfied: (−3) very, (−2) somewhat, and (−1) a little, and three possible responses for Satisfied: (1) a little, (2) somewhat, and (3) very. QoLI z-scores are denoted as very low (0–37), low (38–43), average (44–58), and high (59–77).
The Beck Depression Inventory II (BDI-II) is an updated version of Beck's original 21-item index of depression (Beck et al., 1996). Each item is associated with four possible responses of increasing intensity, with scores of (0) little or no depressive concerns/problems, (1) subjective responses representing mild depressive concerns/problems, (2) subjective responses representing moderate depressive concerns/problems, and (3) severe depressive concerns/problems. The BDI-II score range is 0–63, with the maximum possible score of 63 representing very severe depression. BDI-II scores have been variously categorized using labeled ranges (e.g., minimal [0–13], mild [14–19], moderate [20–28], and severe [29–63]; Halfaker et al., 2011).
Exit Survey
At the conclusion of the final visit, an exit survey (see Appendix C) was administered to assess the participants' subjective impressions of the relative merits of the intervention protocol and the perceived benefits from the individual treatment components. One participant did not complete the exit survey, having missed their final visit, which occurred in the initial transition period to virtual visits (described below). The exit survey, administered by the study audiologist, encouraged participants to provide both positive and negative feedback, with the goal of improving the intervention protocol. The survey began with assessment of the three intervention components: therapeutic sound, LS, and counseling. Participants then were asked seven open-ended questions regarding their impressions of each component, their impressions of the overall intervention, any perceived change in their quality of life, their recommendations for improvement of the study protocol, and any other feedback they wanted to provide. Their responses were recorded verbatim by the study audiologist. Participants were then asked to rate the benefit of each of the three intervention components using the following scale: (1) not beneficial; (2) slightly beneficial; (3) moderately beneficial; (4) very beneficial; and (5) extremely beneficial. Finally, participants were asked to rate their hyperacusis condition posttreatment relative to pretreatment using the following scale: (1) much worse; (2) slightly worse; (3) no change; (4) slightly better; and (5) much better.
Analysis Plan
Primary analyses used linear mixed-models implemented in a statistical software package (SPSS Statistics 27, IBM) to evaluate longitudinal change in treatment effect across multiple visits for each study outcome. Linear mixed-model analyses also evaluated slope inequalities between participants for each outcome measure. Regression analyses included evaluation of scores from each sound-based questionnaire as predictors of the output-limiting LS threshold settings.
Secondary analyses using paired-samples t tests also evaluated change in the LS threshold settings, each of the study questionnaire scores, and the categorical loudness judgments between T0 and T6. Additionally, independent-samples t tests evaluated differences in the output-limit LS threshold settings between the TE and non-TE across visits. Statistically significant results were based on p values ≤ .05. Cohen's d statistic (Cohen, 1988) and the associated 95% confidence intervals (95% CIs) also were calculated to assess the practical significance of treatment-related change in each outcome measure, specifically, the effect size of the change between T0 and T6 (or final visit).
Results
Primary Outcome
Our primary outcome measure, LS release in dB for the designated TE, is shown numerically in Table 6 across follow-up visits T1 to T6 (relative to T0) for the group and for each participant. The (absolute) LS settings across study visits are shown graphically in dB SPL in the left panel of Figure 2. The normalized-to-baseline (relative) LS change settings, representing the LS release values from Table 6, are replotted in the right panel of Figure 2. These absolute values in the left panel represent levels corresponding to the loudness judgments for Category 6 running speech over the course of the intervention. The relative values in the right panel represent the change in these judgments relative to baseline values. To reiterate, incremental LS release over the course of the intervention indicates progressively less reliance on output limiting, consistent with improved sound tolerance. Accordingly, larger values in Table 6 and in Figure 2 indicate greater improvements in a participant's hyperacusis condition over the course of the intervention.
Table 6.
Average and individual increases in output-limit levels (LS release in decibels) in the designated test ear at each treatment visit, T1–T6, and at the final visit relative to the respective baseline setting at T0.
| Participant | Test Ear |
Treatment visit |
||||||
|---|---|---|---|---|---|---|---|---|
| T1 (n = 12) |
T2 (n = 12) |
T3 (n = 12) |
T4 (n = 12) |
T5 (n = 10) |
T6 (n = 10) |
Final (n = 12) |
||
| 1 | R | 11 | 14 | 29 | 33 | 33 | 33 | |
| 2 | R | 1 | 1 | 41 | 51 | 51 | ||
| 3 | L | 16 | 26 | 26* | 30 | 34* | 42 | 42 |
| 4 | L | 4* | 4* | 22 | 27* | 29 | 31 | 31 |
| 5 | R | 0 | 3 | 18* | 18* | 21* | 21 | |
| 6 | L | 0 | 15 | 15 | 15 | 15 | 16 | 16 |
| 7 | R | 3 | 27 | 12 | 12 | 14 | 47 | 47 |
| 8 | R | 30 | 20 | 27 | 30 | 45 | 68 | 68 |
| 9 | R | 4 | 4 | 4 | 4 | 4 | 4 | 4 |
| 10 | L | 28 | 28 | 28 | 31 | 34 | 41 | 41 |
| 11 | L | 5 | 13 | 16 | 30 | 36 | 36 | 36 |
| 12 | L | 10 | 10 | 15 | 18 | 26 | 27 | 27 |
| Average | 9 | 14 | 21 | 25 | 27 | 33 | 35 | |
Note. Final visit denotes the last measurement available for each participant. Asterisk (*) indicates adjustments made at virtual visits and blank cells indicate missed visits. The total number (n) of individuals with available data is shown for the designated visit. R = right; L = left; T1–T6 = treatment visits 1–6; LS = loudness suppression.
Figure 2.
Absolute and relative LS activation threshold settings for each participant (gray symbols and lines) and the average (bold black square symbol and lines) for the TE as a function of study visit. Left panel: Absolute LS activation threshold settings programmed in the protective devices in dB SPL as a function of study visits T0–T6 and for the final visit. The number of participants contributing to the average at each visit includes all available data for that visit (see Table 6). The final value is the average of the final visit data available for each of the 12 participants and for the group (two participants did not have a T6 visit). Right panel: Individual and group normalized change in the LS setting, release of LS in dB, across visits T1–T6 and final values relative to the baseline setting measured at T0. All absolute and relative values shown in both panels are based on and are synonymous with the participants' threshold categorical loudness judgments for “loud, but OK” running speech and the corresponding normalized judgments. dB SPL = decibels in sound pressure level; LS = loudness suppression; Re. = relative to; T0–T6 = study visits.
The T0 output-limiting LS settings across the 12 participants in Figure 2 ranged from low values of 35 (minimum device setting) up to 68 dB SPL; corresponding trial-ending values ranged from 39 to 115 dB SPL (maximum device setting). Group mean changes in the LS settings across the six follow-up visits were evaluated in linear mixed-model analyses. The range of T0 values differed markedly across participants, reflecting different levels of hyperacusis severity. The fixed effects yielded an estimated LS setting at T0 of 47 dB SPL; these settings were significantly different between participants at T0 (β = 47, SE = 4.0, p < .001).
The average LS setting was almost 82 dB SPL at the final visit, with an average statistically significant increase in the LS setting of 5.6 dB per visit (β = 5.64, SE = 0.81, p < .001) between the T0 and final visits. There were no significant slope differences among the participants. The mean LS release between the T0 and final visits was 34.8 dB (SE = 4.91). This average improvement in sound tolerance between the T0 (average = 44.9 dB SPL) and final visits (average = 79.7 dB SPL), evaluated with a paired t test, was significant (t = −7.08, df = 11, p < .001). Time of treatment also was significantly associated with higher output-limit threshold settings (β = 5.29, SE = 0.97, p < .001, R2 = .28).
The resulting individual improvements in sound tolerance over the course of the trial ranged from minimum LS release of 4 dB for Participant 9 to maximum LS release of 68 dB for Participant 8. Eleven of the 12 participants had a positive change in their LS settings of at least 16 dB over the course of the trial, and nine of the 12 participants attained their largest single-visit changes in the LS settings across visits T1–T3.
We also measured LS release independently for the non-TE of each participant over the course of the trial. Paired t tests comparing LS release between the TE and non-TE yielded nonsignificant differences (p > .05) at each follow-up visit. The mean change in the LS settings of 31.2 dB (SE = 2.97) between the T0 and final visits was highly significant for the non-TE (evaluated in a paired t test: t = −10.5, df = 11, p < .000).
Thus, the analyses of our primary outcome measure are consistent with the study hypothesis stated in the first paragraphs. Namely, we hypothesized that the transitional intervention would affect a statistically and clinically meaningful improvement in sound tolerance. Indeed, the treatment effects were remarkably large. Their practical significance is highlighted in an analysis of Cohen's d (Cohen, 1988) values for the TE and non-TE. The respective effect sizes were significant and huge for the mean release of LS between the T0 and final visits (greater than 2 SDs [Sawilowsky, 2009] for both the TE, d = −2.045, 95% CI [−3.046, −1.016], and the non-TE, d = −3.033, 95% CI [−4.393, −1.651]). Moreover, because the intervention similarly and independently affected LS release in the TE and non-TE over the course of the trial, these results for our primary outcome measure support the idea that the effect of the treatment is, in fact, one of recalibrating hyper-gain central auditory processes. Thus, this bilateral treatment-driven recalibration effect is consistent with reduction in elevated maladaptive central neural activity, which is believed to give rise to LH (Henry, 2022). The resulting restoration of normal central auditory gain function, with homeostatic balance between excitatory and inhibitory central neuronal processes, is our presumptive intervention mechanism for resolution of LH.
Categorical Loudness Judgments
Except for the virtual visits denoted in Table 6, threshold categorical loudness judgments for “loud, but OK” (Contour Test Category 6) were measured in the laboratory for all participants. These loudness judgments were measured under headphones at each study visit from T0 to T6 for pulsed warble tones presented at 500 and 2000 Hz. Additionally, “comfortable” loudness judgments for Contour Test Category 4 and threshold loudness judgments for Category 6 also were measured for spondee stimuli at T0, T3, and T6. (The Category 4 judgments were considered only for determining comfortable levels for subsequent word recognition test measurements.) The average change between visits T0 and T6 revealed relatively small (insignificant) increases (~5–10 dB depending on the stimulus condition) in the threshold loudness judgments for Category 6.
Contour Test loudness judgments for the six categories from Category 1 (“very soft”) up to the threshold judgment for Category 6 also were measured for a 4000-Hz pulsed warble tone. The resulting group average loudness judgments (in dB HL) measured at T0 and at the final visit for each participant are shown in Figure 3 as a function of Contour Test category for the respective study visits. As expected, based on the results above, there was a consistent trend for the categorical loudness judgments to shift to higher levels at the final visit relative to those measured at T0. However, again, none of the increases in the mean judgments between the T0 and final visits reached statistical significance in paired t tests of the respective sets of judgments for any Contour Test category or in the corresponding estimates of effect sizes estimated by Cohen's d values. Accordingly, these statistical analyses are not presented here.
Figure 3.
Average categorical loudness judgments in dB HL measured at T0 (dashed function) and the final visit (solid function). The loudness judgments for Contour Test categories 1 (very soft) through 5 (comfortable but slightly loud) are the median values for each category, whereas the loudness judgments for category 6 (loud, but okay) represent threshold values for that category. dB SPL = decibels in sound pressure level; T0 = baseline visit.
Thus, on average, loudness judgments for the warble tone and spondee stimuli increased over the period of the transitional intervention. These treatment-driven changes, however, were small (typically < 10 dB) and not statistically significant for our group of study participants.
Speech Tests (Word Recognition and HINT Scores)
Word recognition and HINT measurements were performed in sound field with the treatment device in use in the TE at visits T0, T3 and T6 (the non-TE device was deactivated). The tests were administered at presentation levels based on each participant's measured Category 4 and threshold Category 6 loudness judgment levels for spondees. The NU-6 scores changed from 97.3% (Category 4) and 98.3% (Category 6) at T0 to 99.5% and 99.3%, respectively, at T6. Thus, excellent initial performance created a ceiling effect for which no significant changes in these measurements were observable. The HINT was administered with the background noise fixed at the Category 4 or threshold Category 6 loudness judgment levels, while the speech presentation level was varied adaptively. The average change in SNR from baseline to T6 was from −4.2 to −2.2 for Category 4 and from −5.8 to −5.9 dB for Category 6. Like the NU-6 scores, no significant changes in the SNR values were measured for the HINT despite the higher presentation levels at T6.
Questionnaire Responses
The responses to all study questionnaires were collected from each participant at laboratory visits T0, T3, and T6, and at all virtual visits. Below we initially consider the global scores for each sound-based questionnaire denoted in Figure 4. These scores are shown for each individual and for the group mean as a function of the respective study visit. Cutoff and category range scores (indicating a hyperacusis condition or degree of severity) for each questionnaire are inset in the respective panels. These values are based on original validation studies or other documentation provided by the developer (except for the NAQ for which no severity categories have been defined). Lower scores indicate improvement in the hyperacusis condition.
Figure 4.
Individual and group mean scores on the four sound-based questionnaires and on the two non-sound–based questionnaires. The inset legend in each panel denotes ranges of scores (where available) for use in designating levels of severity for that measure. These ranges, categories, and cutoffs are based on the recommendations of the original developers of each questionnaire. BDI-II = Beck Depression Index-II; GÜF = English translation of German Index of Sound-Related Distress; HQ = Hyperacusis Questionnaire; NAQ = Noise Avoidance Questionnaire; QoLI = Quality of Life Inventory; TS-H = Tampa Scale for Hyperacusis (modified from Swedish version).
All sound-based questionnaires in Figure 4 reveal the same general pattern, which is highlighted in the group results. Namely, scores at T3 and T6 (and for the final visit mean value) are lower than the baseline score at T0. Except for the TS-H, for which the group scores were still decreasing at T6, the corresponding scores for each of the other questionnaires approached or were at asymptotic performance by T3. Mixed-model analyses, shown in Table 7, indicated that individuals improved significantly (p < .05) by ~1.7 points on HQ, ~1.8 points on GÜF, ~2.1 points on TS-H, and ~ 3.4 points on NAQ per visit across the six-visit follow-up period. Regression analyses also revealed that time of treatment was significantly associated with lower HQ scores (β = −1.52, SE = 0.6, R2 = .18, p = .01), lower GÜF scores (β = −1.69, SE = 0.5, R2 = .25, p = .003), and lower TS-H scores (β = −1.93, SE = 0.6, R2 = .24, p = .004). Time of treatment was not significantly associated with lower NAQ scores (β = −2.7, SE = 1.7, R2 = .071, p = .128) perhaps because large portions of time were spent in isolation within the home during the COVID-19 lockdown period; therefore, change in sound avoidance may have been problematic for some participants to gauge. Additional regression analyses assessed whether each of the sound-based questionnaire scores was a significant predictor of the treatment-driven output-limiting LS settings. In contrast to the nonsignificant time-of-treatment effect for the NAQ, it was the strongest predictor of change in the LS settings (β = −0.45, SE = 0.13, R2 = .25, p = .002), with 25% of the variability explained by the model, followed by TS-H (β = −0.20, SE = 0.06, R2 = .22, p = .004), then HQ (β = −0.16, SE = 0.06, R2 = .17, p = .012), and, finally, GÜF (β = −0.13, SE = 0.06, R2 = .13, p = .029).
Table 7.
Mean questionnaire scores (and standard deviations) at T0, T3, T6, and final visit along with the results for mixed-model analyses, paired t tests, Cohen's d values (with 95% confidence intervals), and corresponding effect sizes.
| M (SD) | M (SD) |
M (SD) Mixed-model analysis |
M (SD) Paired t test |
Cohen's d [lower CI, upper CI] |
Interpretation (effect size) | |
|---|---|---|---|---|---|---|
| T0 n = 12 |
T3 n = 10 |
T6 n = 9 |
Final n = 12 |
Pair (T0, Final) |
Pair (T0, Final) |
|
| HQ | 30.6 (6.2) |
22.3 (8.0) |
21.7 (9.4) β = −1.7, SE = 0.39, p < .001 |
21.1 (9.1) t11 = 4.737, p = .001 |
1.367
[0.55, 2.15] |
Very large |
| GÜF | 27.0 (7.7) |
17.8 (6.2) |
17.1 (7.4) β = −1.76, SE = 0.39, p = .001 |
16.9 (6.83) t11 = 4.427, p = .001 |
1.278
[0.49, 2.04] |
Very large |
| TS-H | 44.3 (8.1) |
36.4 (7.9) |
32.98 (10.2) β = −2.07, SE = 0.38, p = .001 |
33.0 (9.9) t11 = 5.987, p < .001 |
1.728
[0.80, 2.62] |
Very large |
| NAQ | 56.9 (19.7) |
39.7 (25.7) |
41.3 (27.2) β = −3.37, SE = 1.05, p = .003 |
39.0 (26.2) t11 = 2.731, p = .020 |
0.788
[0.12, 1.43] |
Large |
| QoLI | 33.3 (18.0) |
38.8 (15.0) |
39.2 (14.9) β = 0.8, SE = 0.53, p = .14 |
38.1 (13.8) t11 = −1.389, p = .192 |
−0.401 [−0.98, 0.19] |
Small |
| BDI-II | 15.5 (7.7) |
12.8 (9.0) |
11.8 (9.3) β = −0.63, SE = 0.34, p = .09 |
11.6 (8.4) t11 = 1.972, p = .074 |
0.569 [−0.05, 1.17] |
Medium |
Note. Bold values indicate statistically significant results. T0, T3, T6 = study visits; CI = 95% confidence interval; SE = standard error; HQ = Hyperacusis Questionnaire; NAQ = Noise Avoidance Questionnaire; GÜF = English version of the German Index of Sound-Related Distress; TS-H = Tampa Scale for Hyperacusis; QoLI = Quality of Life Inventory; BDI-II = Beck Depression Inventory–II.
Paired t-test comparisons of the mean scores measured at T0 and at the final visit (shown in Table 7 along with the mean and standard deviation values at T3 and T6 for each questionnaire) yielded statistically significant (p < .05) improvements for each sound-based questionnaire. Corresponding effect sizes, indexed by Cohen's d values, also were significant for each of these questionnaires (see Table 7).
The response scores for the non–sound-based questionnaires at T0, T3, and T6 and for the final visit also are shown graphically for each of the individual participants and for the group in Figure 4. The group means and corresponding standard deviations for these study visits are also shown numerically in Table 7. Although there is a trend for the BDI-II scores and some of the QoLI scores to improve across study visits, the treatment-related effects measured in the mixed-model analyses, regression analyses, and paired t tests were nonsignificant (p > .05). Moreover, none of the effect sizes were significant for either of these questionnaires.
Exit Survey
The exit survey was completed by 11 of the 12 participants at their T6 visit. The survey assessed their degree of satisfaction with the intervention protocol and benefit from the intervention components. Participant 1, who did not complete the T6 visit, did not complete the exit survey. Except for one participant (Participant 9), all participants rated the intervention components as beneficial. In fact, the proportion of participants who responded “very” or “extremely” beneficial was 82% for counseling, 82% for sound therapy, and 64% for LS. In response to whether there was an overall change in the hyperacusis condition at end of treatment relative to pretreatment, eight of the nine participants who responded indicated that their hyperacusis conditions were “slightly” or “much better,” with Participant 9 responding “no change” at the conclusion of treatment. Ten of the 11 participants, when asked about their quality of life when using their intervention devices, reported improvement (Participant 9 denied improvement). The benefit described most frequently was their relief from planning sound avoidance strategies, thus indicating the participants' reduced concerns about exposure to offending sound levels throughout the intervention. This benefit, in turn, enabled increased participation in normal daily activities. Individual participant reports of subjective satisfaction with the fit and performance of their intervention devices indicated that the devices were comfortable and reliable for all-day usage. Only one participant, with a history of autism spectrum disorder and sensory issues, expressed dissatisfaction with device comfort and retention associated with the earpieces. These concerns were resolved using standard techniques (i.e., through a combination of physically modifying the earmold and use of adhesive tape to maintain the device in the preferred position behind the ear). Participants stated that counseling was highly beneficial in explaining their condition, the purpose of the intervention, and allowing them to share and communicate their condition with others. Notably, 100% of participants indicated they would encourage other hyperacusis sufferers to participate in the intervention protocol. When participants were asked for specific recommendations to improve the protocol, the most common response (n = 4) was to extend the duration of the treatment or allow the devices to be kept permanently, followed by no recommendations for improvement (n = 3), get the product to market (n = 1), offer a rechargeable device (n = 1), and improve the physical comfort of the occluding ear plugs (n = 1).
Discussion
Successful Transitional Approach
The outcomes from this trial were highly significant statistically and clinically based on sizable treatment-related change in the primary outcome measure (i.e., LS release) and on meaningful change on those self-report questionnaires that addressed hyperacusis- and sound-related problems and concerns of our participants (i.e., hypersensitivity to sound, sound-related distress, annoyance, pain, and avoidance). Indeed, based on these objective and subjective metrics, participants' anecdotal reports, and their positive responses to our exit survey, our transitional intervention was highly successful and beneficial for all but one of the 12 participants who completed the protocol. Even the nonrespondent participant (Participant 9), who was noncompliant with the intervention protocol, was motivated to continue in the study rather than withdraw. Moreover, the transitional intervention appears broadly applicable based on the diverse set of causal associations and comorbid conditions among our positive respondents, including six who reported tinnitus. Our sample also included a broad range of hyperacusis severity, indexed by baseline LDL judgments spanning atypically low values from 30 to 75 dB HL. Thus, the positive intervention effects across multiple relevant outcome measures from a diverse sample of hyperacusic individuals, representing a breadth of etiologies and severity, indicate that our approach is likely appropriate for a broad swath of individuals with hyperacusis complaints and related conditions of decreased sound tolerance. We may therefore infer that the mechanism that gives rise to LH across our diverse sample is, indeed, attributable to neural hyperactivity within the central auditory pathways and this maladaptive hyperactivity can be downregulated to reduce or eliminate LH and associated distress and stress responses with our transitional intervention. That one participant with pain hyperacusis was unsuccessfully treated with our intervention may indicate that a different mechanism or pathway mediates this condition. We considered this possibility in our counseling protocol and shared a visual aid to highlight that the pathway mediating pain hyperacusis is likely different from that for LH (see Cherri et al., 2024, for a deeper discussion).
These positive treatment-related outcomes are especially impressive for our primary outcome measure of LS release, for which the average change from baseline at the end of the trial period still had not plateaued after achieving more than 34 dB of improvement in sound tolerance for running speech. This lack of plateau raises the intriguing possibility that the observed treatment effects were not maximum treatment effects and that continuation in the intervention protocol for a period longer than 6 months may continue to improve outcomes for some participants. Thus, our transitional intervention was successful by almost all relevant metrics of measurement used in this trial for almost all participants completing the trial.
Contributions of Counseling, Sound Protection, and Therapeutic Sound
Our trial design and limited sample size do not enable us to dissociate the effects of counseling from the effects of therapeutic sound, but both counseling and sound therapy were considered important components of the intervention as revealed by the participants' responses in the Exit Survey. We also have average estimates of relative use of the two sound therapy options implemented with (ear plug) and without (open dome) the LS activated over the course of the trial period. These estimates, shown in Figure 5, are based on documented usage of the two options reported by the participants at each follow-up visit. Sound generator usage with protective LS was primarily used at the start of the trial, over the first 2–3 months of the intervention, but afterwards, sound generator use without LS was the overwhelmingly dominant form of therapeutic sound adopted by the participants over the remainder of the intervention period. This is not a surprising result since the participants' sound tolerance progressively improved over the course of the intervention, with systematically less need for sound-limiting protection outside of the home. It is noteworthy, however, that most of the participants achieved their greatest improvements in sound tolerance, as documented by LS release and the sound-based questionnaire scores, during the early period when protected sound therapy was primarily used. This is a reassuring result because there was concern a priori that the therapeutic benefits from the sound generators might be limited or retarded by the potentially confounding effects of LS early in the treatment period when the greatest amounts of LS were being imposed (i.e., the output-limiting compression threshold of the protective treatment device was set to respond to the lowest activation levels). This was a critical period of the intervention when the participants' sound tolerance was worst and their exposure to a wide range of healthy sound levels, which is necessary for developing and maintaining a typical dynamic range, was most restricted by LS. Thus, the protective effects of LS, as incrementally released over the course of our transitional intervention, appear to have enhanced rather than stymied the treatment effects from the therapeutic sound generators, especially early on in the trial period. Moreover, responses to our exit survey indicate that the effects of LS were considered beneficial by most of the participants, notwithstanding the potential for fewer opportunities for some of the respondents to use LS outside of the home during the COVID-19 lockdown period.
Figure 5.
Average percentage of usage of open dome and closed (protective) earplug configurations with sound generators as a function of treatment visit.
It is important to reiterate that we considered the effects of LS to be protective rather than therapeutic. However, Susan Gold (The Hyperacusis Network Message Board, April 8, 2009; under Report from Audiology Convention & Hyperacusis) has suggested that strategies that reduce reliance on counterproductive sound-limiting protection should be considered therapeutic in the treatment of hyperacusis. Gold lays out her argument as follows: “The goal is to wean each patient off overprotection at a rate that is comfortable for them. Sometimes it involves using less and less protection before any sound generators are implemented. For example, foam earplugs can be cut longitudinally so the ear canal is less filled, or musician's earplugs can be fit starting at a 15 dB filter and moving to 9 dB and then 3 dB. Reducing the noise protection is in itself a form of sound therapy.” Thus, by Gold's logic, LS release (in and of itself) may be considered therapeutic for LH in that it reduced reliance on counterproductive sound-limiting hearing protection over the course of our trial while promoting greater exposure to safe and healthy sound levels and improved sound tolerance. Obviously, a separate trial would be needed to establish the therapeutic effects (if any) of LS release in isolation.
Even without release, the LS approach has been shown by Sammeth et al. (2000) to be beneficial for individuals with hyperacusis in some situations. Moreover, the LS strategy, using unity gain, is advantageous in eliminating the counterproductive effects of the earplug, thus enhancing audibility for soft sounds while suppressing potentially offending louder sounds. The net effect of the LS approach (without release) is to provide the individual with LH a larger dynamic range than otherwise would be afforded by use of sound-attenuating earplugs alone. However, LS without release would not appear to offer a therapeutic benefit, and even with release, LS likely requires adjuvant sound therapy (and appropriate counseling) to achieve the meaningful therapeutic effects reported here. Consistent with the idea that LS release is beneficial for a positive treatment benefit, we note that Tyler et al. (2015) and Searchfield and Selvaratnam (2018) have proposed that progressive reduction in output limiting can be used as a management strategy for aided hearing-impaired individuals with hyperacusis.
One last point deserves mention here, namely, the available range of output limiting in this study was appreciably greater than that reported by Sammeth et al. (2000). That is, the lower limit for LS activation in the device described by Sammeth et al. was about 65 dB SPL, whereas the lowest limit for activation of LS in this study was about 35 dB SPL. Thus, this lower limit for LS activation in our device allowed for protective management and transitional treatment of individuals with much more severe hyperacusis conditions than was possible with the device described by Sammeth et al. (2000).
Reconsideration of the Jastreboff Treatment Model for Hyperacusis and Misophonia
Our treatment approach was grounded in many of the concepts and principles espoused by Jastreboff and Jastreboff (2014) in their treatment model for hyperacusis and misophonia. The latter condition, also referred to as “annoyance hyperacusis” (Tyler et al., 2014), has been closely linked to primary hyperacusis in the literature. Indeed, Baguley and Andersson (2008) have proposed that primary hyperacusis will characteristically have an associated distress component; this component may be the comorbid condition described by Hazell et al. (2002), who reported that most of their patients with decreased sound tolerance exhibited symptoms of misophonia. Subsequently, Jastreboff and Jastreboff (2014) related that over 90% of their sample of 201 consecutive patients with decreased sound tolerance had elements of misophonia. Our transitional intervention only superficially considered misophonia during the interview of the participant at the intake visit and in presentation of this concept in the counseling protocol. Rather, we focused on weaning the study participants from self-imposed sound isolation and from overuse of their counterproductive HPDs while offering treatment with enriched sound to resolve their LH. Given the success of our intervention for participants with a wide range of atypically low baseline LDLs, including some as low as 35 dB HL, the Jastreboff model appears problematic in some of its underlying assumptions and warrants re-examination here.
In their model, Jastreboff and Jastreboff (2014) represent primary hyperacusis (i.e., LH) as an abnormal physiological condition of neural hyperexcitability, which gives rise to elevated neuronal gain within the central auditory pathways. The resulting auditory hyper-gain response seemingly is pervasive for virtually all forms of supra-threshold sound stimulation (independent of the meaning or context of the sound). These basic concepts are fundamental in our intervention for LH. Moreover, they propose that in cases of pure hyperacusis, the condition is restricted to a range of LDL values between 60 and 85 dB HL, which is the range of baseline LDL values for half of the participants in this study, with six participants having at least one LDL value < 60 dB HL at either 1000 or 8000 Hz. Jastreboff and Jastreboff (2014) further assume that LDL values below this range reflect some degree of associated misophonia. They suggest that in cases of severe hyperacusis, the associated negative reaction will negatively reinforce and create “a conditioned reflex linking specific sounds with something negative” (p. 112). This is the mechanism in their treatment model for the misophonia presumed to accompany debilitating LH. Jastreboff and Jastreboff (2014) further posit that extremely low LDL values (i.e., 30 dB HL) are indicative of a pure form of misophonia, which is largely independent of the offending sound level. Rather, they propose that pure misophonia represents a negative reaction that is primarily dependent on the meaning, context, or previous exposure of the affected individual to a given sound or class of offending sounds. Thus, it is the negative reaction to these offending sounds that creates a conditioned reflex with reactionary nonauditory processes (within the limbic and autonomic nervous systems) that accounts for misophonia in the Jastreboff model. Therefore, in their model of pure misophonia, the neuronal activity and response to sound within the central auditory pathways are functionally normal and it is the negative reactions to the sound, mediated by heightened neuronal activation within reactive nonauditory processes, that give rise to misophonia.
Accordingly, in the Jastreboffs' neurophysiological model, pure misophonia is mediated by a different and wholly independent mechanism from classical LH. To reiterate, they consider that the latter is principally a hyper-gain neuronal response to virtually all sounds, independent of their meaning or context; the inordinate LH response is confined to the central auditory pathways with no involvement of the reactionary nonauditory processes that give rise to the negative emotional and physiological responses (to the specific class of sounds) that mediate and characterize misophonia. Per Jastreboff and Jastreboff (2014), the LH-related distress to offending sounds results when the hyper-gain activity within the central auditory pathways reaches a level that creates and activates reactionary neuronal connections (i.e., conditioned responses) innervating the limbic and autonomic nervous systems. The strength of this activation is ultimately responsible for the associated negative emotional distress and physiological stress responses to LH. Thus, debilitating LH with associated distress presumably shares with misophonia similar reactionary nonauditory processes in the Jastreboff model. Consequently, these common processes and the associated negative reactions to sound are not necessarily mutually exclusive in debilitating cases of hyperacusis.
Jastreboff and Jastreboff (2014) contend that because the underlying mechanisms that give rise to primary hyperacusis and primary misophonia differ, with auditory pathway gain excessive in the case of the former and functionally normal in the case of the latter, the two conditions will not be responsive to the same treatment approach. Indeed, they state that the misophonic component of decreased sound tolerance cannot be treated with therapeutic sound of the kind we used successfully in our intervention to recalibrate the auditory pathway gains of our LH participants. Instead, they propose a protocol of desensitizing sound exposures to the specific sounds or class of sounds that are bothersome for the misophonia patient.
Only one participant, Participant 9, was unsuccessful with our transitional intervention and he had baseline LDLs as low as 30 dB HL. Participant 9 described symptoms of both pain (his primary complaint) and loudness hyperacusis, presumably with some degree of LH-related emotional distress, physiological stress, or both. His failure with our intervention supports the Jastreboffs' idea that a different treatment strategy was needed for him, ostensibly one focused on reducing his pain response to sound. However, Participant 9 also was the sole participant in our sample who did not adhere to the intervention protocol (perhaps because our intervention approach was in fact an inappropriate treatment strategy for him); he continued daily use of ear muffs and noise-canceling sound-reducing devices throughout the intervention period. In contrast, our intervention-compliant participants, Participants 3, 5, 7, 10, and 12, had LDL judgments that fell below the lower limit (60 dB HL) proposed by the Jastreboffs' for delineating hyperacusis, with at least one of Participant 12's LDL judgments measured at 35 dB HL. The latter participant's success with our intervention, and that of the other compliant responders with low LDL values, in the range from 45 to 55 dB HL, appears to belie some of the Jastreboffs' assertions. Most notable is their assertion that an enriched sound intervention protocol of the kind we have implemented here cannot be used to treat individuals with very low LDLs or the distress-related misophonia that ostensibly is induced by and accompanies severe hyperacusis. Also, it seems unlikely that the counseling delivered in our intervention, designed specifically for recalibrating elevated auditory pathway gain in LH, would have been effective for treating comorbid misophonia. In fact, as noted above, the Jastreboffs' recommend a distinctly different counseling and sound exposure protocol for misophonia from that for treating primary hyperacusis. Here, it is interesting to note that their desensitization protocol for misophonia is not unlike one of the “successive approximations” approaches proposed by Tyler et al. (2015) for hyperacusis. The latter sound therapy strategy is implemented with counseling in their Hyperacusis Activity Treatment protocol (Tyler et al., 2015, 2022). Their sound therapy approach begins with controlled exposures to recorded low levels of bothersome sounds. Over several weeks of daily exposure, at a prescribed time, the patient listens to these bothersome sounds at increasingly higher levels and over longer durations. Eventually, as the hyperacusis condition improves, the patient moves to locations in which these offending sounds are produced, ostensibly, leading to resolution of hyperacusis for these listening conditions.
It is relevant here to note that, based upon their patients' responsiveness to their misophonia treatment protocol, Jastreboff and Jastreboff conjectured that their misophonia patients are distinctly different from those with a psychiatric condition of misophonia. However, we suppose that our intervention for LH might have weakened or possibly severed the conditioned reflex that the Jastreboffs posit to be responsible for “hyperacusis-induced” misophonia. This disruptive effect, presumably fostered by treatment-driven reduction in the hyper-gain response within the auditory pathways, would also reduce the activation of the reactionary nonauditory processes that LH and misophonia share in the Jastreboff treatment model, thereby reducing the associated negative reactions to LH. Also, as the severity of LH increased, the associated negative reactions to LH would be expected to increase proportionally. Accordingly, the contribution of the counseling component in our intervention would have become increasingly important in habituating the elevated negative reactions to LH (see Cherri et al., 2024). An alternative explanation for the positive treatment effects achieved by our participants with very low LDLs is that the range of LDLs assumed by Jastreboff and Jastreboff (2014) to signify primary hyperacusis is larger than 60–85 dB HL and extends to lower sound levels than they posited.
Thus, the concept of misophonia and the use of this terminology in the hyperacusis literature is complicated, often confusing, and likely will continue to be problematic to delineate from related conditions of decreased sound tolerance for the foreseeable future. Indeed, there currently is no recognized diagnostic code or standardized diagnostic criteria for misophonia (Brout et al., 2018; Dozier et al., 2017). Notwithstanding this confusion and the apparent conflicts with some of the model assumptions above, it is evident that our transitional intervention was successful for participants with varying severity of decreased sound tolerance documented by atypically low baseline LDLs spanning at least a 40-dB range, including values as low as 35 dB HL. Moreover, the hyper-gain basis for LH and its treatment-driven recalibration are entirely consistent with the Jastreboff model for primary hyperacusis; our intervention strategy; and, ultimately, the successful intervention outcomes reported here.
Incongruent Loudness Judgments
Perhaps the most unexpected result in this study was the apparent lack of correspondence between change in the threshold for Category 6 (“loud, but OK”) loudness judgments for running speech (which was our basis for LS release) and that for pulsed warble tones and spondee words over the course of the intervention. Indeed, whereas the former judgments increased on average by more than 34 dB between baseline and end of treatment, the average changes in the Category 6 judgments for warble tones were 7 dB (500 Hz), 10 dB (2000 Hz), and 5 dB (4000 Hz), and that for spondees was only about 6 dB. Some of this incongruity reflects different procedural methods. The former was measured for “real world” running speech in sound field through the participants' protective treatment devices in one or two ascending trials. The latter set of judgments for the tone and spondee stimuli were collected under headphones. Consequently, the participants may have trusted the sound protection from their LS-activated devices in sound field, but not the unprotected sound stimuli delivered via headphones. They therefore were less likely to preempt the loudness judgment process (out of concerns that increasing presentation levels might exceed their tolerance levels) when using their protective devices rather than listening via headphone presentation.
It should be reiterated here that the “loud, but OK” loudness judgments (based on three ascending trials) were based on levels corresponding to Category 6 rather than higher standard median levels within the category as described by Cox et al. (1997). Thus, our loudness judgments, including those for running speech (which also reflect threshold levels for Category 6 judgments), underestimate the median level typically reported for Category 6 per the Contour Test of Loudness (Cox et al., 1997). Even so, the large treatment-related differences in the loudness judgments for running speech with those for pulsed warble tones and spondees in this trial are striking. In fact, these relative intrasubject differences are among the largest we have seen reported in the literature for laboratory stimuli, and these differences are consistent across 11 of 12 study participants. It is, however, well established that loudness judgments for brief tones used in the audiology clinic are not representative of those made under real-world conditions (Altin et al., 2023; Filion & Margolis, 1992; Jahn, 2022; Munro & Patel, 1998; Punch et al., 2004; Sheldrake et al., 2015; Zaugg et al., 2016). Our results also indicate that loudness judgments for brief tones and simple word stimuli do a poor job of representing loudness judgments for running speech in sound field. Moreover, it is also well established that individuals with misophonia may produce LDLs and related measures of sound tolerance almost anywhere within the auditory dynamic range for loudness, including from 30 to 120 dB HL depending on comorbidity with LH (Jastreboff & Jastreboff, 2014; Sheldrake et al., 2015). This issue further complicates the use of sound tolerance measures as diagnostic indices for the various forms of hyperacusis recognized by Tyler et al. (2014), especially if misophonia is in fact comorbid in many individuals reporting decreased sound tolerance (Hazell et al., 2002; Jastreboff & Jastreboff, 2014). Nevertheless, the consistent and large incremental release of LS across the participants in our trial, whose LDLs represented at least a 40-dB range of atypically low loudness judgments, bodes well for applications of our transitional intervention for LH and, more importantly, for the generality of our intervention for treating a broader population of individuals with diverse etiologies and generic symptoms of decreased sound tolerance.
Returning to the treatment-related changes in the Category 6 threshold judgments, these changes for the warble tone and spondee stimuli are in line with, albeit somewhat smaller than, the standard Category 6 loudness judgments reported by Formby et al. (2015) for 11 mildly hyperacusic (average LDLs ~85 dB HL) hearing-impaired individuals. The latter were treated over a similar period to that of our intervention period (~6 months). Moreover, they were treated with a closely related treatment strategy that used a comparable counseling protocol with similar therapeutic sound generators (without LS). Thus, stimulus duration appears to be the most parsimonious explanation for the sizable differences in the loudness judgments between the long-duration running speech stimulus and the brief-duration tone and spondee stimuli evaluated in our trial. Here it is noteworthy in their Figure 7 that Formby et al. (2015) showed progressively larger treatment-driven incremental changes over the course of their intervention and these treatment effects increased as a function of increasing loudness category. After 7 months of treatment, their sound-sensitive hearing-impaired listeners achieved posttreatment “uncomfortably loud” judgments (Contour Category 7) averaged for their 500, 2000, and 4000-Hz warble tone conditions that were several dB greater than those for either their “loud” (Contour Category 5) or “loud, but OK” (Contour Category 6) judgments. Thus, if we were to add a similar treatment effect of 3–4 dB at treatment end to our participant's average threshold loudness judgments for Category 6, then the resulting values would agree well with the statistically significant treatment-driven improvements in the “uncomfortable loudness” judgments reported by Formby et al. (2015).
Although not directly comparable, it is informative to compare the treatment-related changes in our Category 6 threshold loudness judgments for the warble tone stimuli with corresponding treatment-related changes reported in two relevant reports in the literature, one based on a counseling intervention, cognitive behavioral therapy (CBT; Jüris et al., 2014), and the other based on a novel therapeutic sound intervention without counseling (Noreña & Chery-Croze, 2007). The CBT treatment–related LDL changes described by Jüris et al. (2014; for hyperacusis patients treated with six sessions of CBT over a 2-month period) were like those reported above for our Category 6 threshold judgments. However, only 57% of their patients were responsive to CBT at treatment end. It is noteworthy that their CBT protocol called for progressively increasing exposure to sound over the course of the intervention. Our treatment-driven changes for Category 6 threshold judgments and those of Jüris et al. for LDL judgments are roughly half the size of the treatment-related LDL changes reported by Noreña and Chery-Croze (2007) for a group of hearing-impaired tinnitus patients with hyperacusis. The latter investigators described an intervention that used a dynamically changing stimulus, tailored to the high-frequency range of primary hearing loss, which was delivered under headphones for 1–3 hr a day. Their primary treatment effect was attained after 15 weeks of treatment. In contrast to the treatment effects achieved with CBT and the larger effects reported by Formby et al. (2015), both of which were sustained (in whole or part) for at least a year posttreatment, the treatment effects reported by Noreña and Chery-Croze were not sustained after the treatment concluded. None of the studies noted above achieved a treatment effect that was half of that indexed by LS release in this study, which averaged almost 35 dB. We do not know, however, whether this treatment effect was sustained by our study participants after treatment end.
Lastly, in this context, it is worth considering whether diagnosis and prediction of hyperacusis, based on analyses of patient's LDL judgments and related categorical judgments of sound tolerance, have value for broad identification of individuals with hyperacusis and their remediation. We know that many extraneous variables and confounding conditions (some of which were noted above) affect these judgments (see Punch et al., 2004; Tyler et al., 2014). These factors (e.g., the degree of hearing loss and the slope of the audiogram; Hawley et al., 2007), the instruction set, and patient concerns that the test will cause pain or discomfort (Sheldrake et al., 2015), in turn, influence diagnoses and determination of treatment efficacy for hyperacusis and for related conditions of decreased sound tolerance. Sensitivity and specificity analyses of LDL judgments among hyperacusis patients highlight the significant challenge of utilizing these measures for diagnosis and for prediction of treatment success (Goldstein & Shulman, 1996; Hawley et al., 2007; Sheldrake et al., 2015). Consider Sheldrake et al.'s (2015) analyses of 381 patients with primary complaints of hyperacusis. They reported that particularly low LDL judgments, below 70 dB HL, were highly specific for hyperacusis in their sample. Moreover, they reported that a cutoff for LDL judgments below ~100 dB HL correctly classified 90% of the individuals in their sample with hyperacusis. However, the downside of adopting this latter (lax) cutoff was that the corresponding false alarm rate was between 40% and 50%, which obviously is unacceptably high. Predictions of treatment-related remission of hyperacusis, based on models that input pretreatment LDL judgments (together with audiometric threshold data), are similarly challenged, with the best logistic regression model reaching about 80% (for acceptable false alarm rates; Hawley et al., 2007). Thus, traditional clinical judgments of sound tolerance have limitations. Perhaps our protocol using loudness judgments for “real world” running speech, which was the basis for LS release in this trial, will offer advantages for assessing decreased sound tolerance and evaluating treatment-related change.
Consideration of the Self-Report Outcomes (Questionnaires)
Four sound-based questionnaires were selected for use in this study, including the HQ (Khalfa et al., 2002) and GÜF (Nelting & Finlayson, 2004). These latter instruments are probably the two most widely used self-report tools globally, as well as the two most reported in the literature, for categorizing and quantifying complaints of sound sensitivity, decreased sound tolerance, or hyperacusis-related issues. The stated purposes of the HQ are to quantify and characterize hypersensitivity to sound, whereas those of the GÜF are to assess sound-induced distress/annoyance, providing indices to inform treatment needs and planning for remediation of hypersensitivity to sound. Several reports and critical reviews have addressed the strengths and shortcomings of these two instruments and their respective subscales (Altin et al., 2023; Fackrell et al., 2015; Fackrell & Hoare, 2018; Fioretti et al., 2015); because of reported limitations and confounding of some of the items and subscales (affecting their interpretation), we only considered the respective global scores in our analyses. In this study, we also included the NAQ (Blaesing & Kroener-Herwig, 2012), an instrument designed to assess sound avoidance behaviors and associated anxiety; these latter issues are of obvious relevance and represent a set of problems that our transitional intervention was designed to mitigate. Fackrell and Hoare (2018) state that the NAQ is the English version of the German GÜF (Nelting & Finlayson, 2004), but if so, then a review of the respective formats and the English translation of the items suggests otherwise (Bläsing et al., 2010). The final instrument, the TS-H, was a questionnaire adapted from the pain literature for purposes of quantifying subjective complaints related to sound-induced pain and discomfort (Jüris et al., 2014); we translated the latter Swedish language questionnaire into our English version of the TS-H (Formby & Eddins, 2019). These instruments, with exception of the TS-H, were developed for and validated in tinnitus populations, which potentially confounds the interpretation of some of the questions and associated responses. We are not aware that either the NAQ or TS-H, including the original Swedish version of the latter, has been validated in any language to date.
To our knowledge, no hyperacusis questionnaire has yet been developed and tested for the specific purpose of documenting treatment-related change and efficacy; however, at least two of our instruments, the HQ and TS-H, have previously been used for this purpose. Nevertheless, we found that all four of these questionnaires were sensitive to treatment-related change in this study. Indeed, the average change in score between the T0 and final visits was 9.6 (HQ), 10.1 (GÜF), 11.3 (TS-H), and 17.9 points (NAQ). These changes were statistically and practically significant, including significant correlations with the treatment-driven changes in the participants' LS settings. These group change scores therefore represent meaningful treatment-related improvements over the course of the transitional intervention that are clinically relevant. Moreover, individually, almost all the participants showed meaningful treatment-related improvements on each of the sound-based questionnaires. In fact, all 12 participants achieved treatment-related improvements in their HQ scores, with nine of the participants' HQ scores at their final visit falling in the “No Hyperacusis” range designated by Khalfa et al. (2002). Similarly, 11 of the 12 participants' sound-related distress and annoyance complaints were diminished per their GÜF scores at their final visit. All 12 participants' sound-related pain concerns were diminished at their final visit on the TS-H, whereas the sound avoidance behaviors for nine of the 12 participants were reduced at their final visit per their NAQ scores. Interestingly, the latter questionnaire achieved the strongest and most significant association with the participants' incremental changes in their output-limiting LS settings, supporting our hypothesis that sound avoidance behaviors would be reduced over the course of a successful intervention.
These favorable self-report outcomes are complimented by anecdotal and subjective impressions of interventional benefit and satisfaction reported in the exit survey by most participants. Indeed, the consensus among our participants was that their hyperacusis conditions were “much better” relative to that at onset of the intervention. This collection of positive subjective outcomes and impressions, in turn, mirrors the treatment-driven functional changes in LS release (and the lesser but positive changes in the loudness judgments for brief tone and spondee stimuli) measured in this interventional trial.
The only questionnaire measures in this study that were not significantly changed over the course of this trial were the non–sound-based measures, the QoLI (Frisch et al., 2005) and BDI-II (Beck et al., 1996). We were somewhat surprised by the lack of significant change in the former, but perhaps less so in the latter. We suspect both the QoLI and BDI may have been too generic and nonspecific in their question structures and formats to reveal (sound-specific) treatment-related change relevant to our intervention. To wit, consider the results from the exit survey. When participants were specifically asked about the impact of the intervention on their quality of life, their responses were overwhelmingly positive, with only one respondent reporting no change. Our conjecture here also is consistent with a theme put forth by Tyler et al. (2020), who questioned whether existing quality-of-life questionnaires offer sufficient specificity to assess adequately the effects of hearing-related conditions and associated treatments on an affected respondent's life quality. Perhaps more salient here were the untimely and adverse effects of conducting this interventional trial during an ongoing period of societal malaise, fear, and isolation, beginning at the onset of and spanning the peak periods of the COVID-19 outbreak in the United States. Undoubtedly, these untoward conditions contributed to and confounded participant responses to questions concerning their overall quality of life and mental well-being.
Limitations
The average change in our primary outcome measure between baseline and end of intervention was much larger than any related outcome measure based on treatment-driven change in sound tolerance known to us in the hyperacusis literature. We were surprised by this very sizable treatment-driven change, which, as noted above, mirrored subjective reports of improvement in the hyperacusis conditions of the study participants. While we are aware of some individual examples of such large changes in sound tolerance measures in response to tinnitus retraining therapy (TRT) and related interventions using counseling together with therapeutic sound generators, the largest group treatment-related tolerance changes previously reported are a little more than 20 dB for persons suffering primary hyperacusis (see Formby et al., 2007); these respondents were typically treated over longer intervention periods than reported here. One wonders how large our treatment effects would have been for a longer intervention period (since the treatment effects for most of our participants, as measured by LS release, had not reached a plateau at their final visit). Perhaps the size of the treatment effects in past studies of primary hyperacusis also would have been greater if the investigators had used a running speech stimulus in their measurements of loudness judgments.
Obviously, our promising outcomes are for a small group of individuals with mostly normal audiometric thresholds and complaints and evidence consistent with primary hyperacusis, LH. They were typically treated for approximately a 6-month period over the course of this trial. These outcomes will need to be replicated in controlled trials with comparison groups using larger samples of individuals with LH (with and without hearing loss). Based on the results in this study, longer intervention periods will be necessary to achieve asymptotic treatment effects for our primary outcome. It will be important to include posttreatment follow-up to assess retention of the primary treatment effects and the course of these effects, both during and following treatment. The latter will be especially important for establishing the need for therapeutic dosing after the conclusion of a formal intervention. To establish the specificity and generality of our intervention, trials ultimately will be needed in other hyperacusis groups, including those with well-defined symptoms as categorized by Tyler et al. (2014), namely, annoyance hyperacusis, fear hyperacusis, and pain hyperacusis. Such trials will almost certainly have to be conducted at multiple sites to achieve sufficient numbers of participants to power the studies. Hopefully, these trials can be conducted under conditions that are not impacted by the confounding effects of COVID-19 (and related strains). The latter required us to alter our planned study protocol, including the scheme for setting LS for participants who were unable to come into the laboratory for their scheduled study visits and device adjustments. Despite these alterations of the study protocol during COVID-19, our modifications appear to have been implemented without adverse impact on either the individual or group outcomes. This is an encouraging result that may benefit the study design and protocol of a future trial and, subsequently, clinical follow-up of the transitional intervention via a remote protocol.
Ironically, the atypical societal conditions associated with COVID-19 isolation may in some cases have been advantageous for some participants in offering them greater opportunities to control and prolong their enriched sound environments and therapeutic sound exposures, uninterrupted, in the safety of their homes. That is, COVID-19 isolation may have allowed some (if not many) of our participants, who otherwise would have been required to face the everyday busy noisy world at the start of the intervention, an opportunity to ease gently into the transitional intervention. Moreover, it is possible that COVID-19 isolation may have reduced the early need for and use of the protected sound treatment devices outside of the home (since most of these participants were likely leaving their homes less often and for shorter periods during COVID-19). Thus, these lifestyle changes may have contributed to an underestimate of the usage of protected sound therapy in comparison with that which one might anticipate under normal non–COVID-19 conditions when most individuals would likely be more active outside of their homes. Consequently, this atypical trial period conducted during COVID-19 isolation may, in some instances, have positively or negatively impacted the study outcomes in ways different from those that might have occurred under usual non–COVID-19 conditions.
Eight of the 18 potential study participants evaluated for this trial had tinnitus. Two of these participants dropped out of the trial at the fitting visit after determining that the protective treatment devices (using the closed stented earpieces) exacerbated their tinnitus. Five of the other six participants with tinnitus were successfully treated and completed the transitional intervention. (Participant 9 was the lone study participant with tinnitus who was noncompliant and unsuccessful with our intervention.) Typically, the protocol for treating patients with both hyperacusis and tinnitus begins with treatment of the former first and then the latter after resolution of hyperacusis (Jastreboff & Hazell, 2004; Jastreboff & Jastreboff, 2000, 2014, 2016). Debilitating reactive tinnitus, however, is potentially a disqualifying condition for the application of our full intervention for LH, at least early on when protected sound therapy with occluded ear canals may be expected to exacerbate the perception of tinnitus.
Finally, this trial included mostly individuals with little or no significant audiometric hearing loss. However, the principles and concepts that are the foundation of our transitional intervention appear applicable for hearing-impaired persons with or without LH since both will have reduced dynamic ranges. Indeed, Participant 4, who qualified for participation in this trial with a mild-to-moderate hearing loss, achieved a treatment-driven change in his LS settings of 31 dB between T0 and T6. Moreover, a similar intervention strategy, including the counseling and sound therapy protocols, has been used successfully to improve the sound tolerance and expand the auditory dynamic ranges of hearing-impaired persons with and without hyperacusis (Formby et al., 2015, 2017; Gold & Formby, 2017). Thus, we would expect our interventional approach to be generally applicable for persons with elevated audiometric thresholds (due to sensorineural hearing loss) and those with decreased sound tolerance, most notably LH.
Conclusions
The transitional intervention described in this report is a rigorous and technically challenging protocol that has been vetted through NIH peer review, the U.S. patent office (Eddins et al., 2020), and now in the successful proof-of-concept trial highlighted here. A transitional intervention of this kind has long been needed to reduce the patient's reliance on counterproductive sound protection while at the same time offering therapeutic treatment to resolve the debilitating effects and symptoms of primary hyperacusis. The counseling component of our intervention, described by Cherri et al. (2024), follows a checklist format and structured protocol. The basic concepts presented in the hyperacusis counseling will be familiar to clinicians who have been trained to offer TRT and possibly other structured intervention protocols incorporating sound therapy and related principles for treatment of hyperacusis (e.g., Hyperacusis Activities Treatment; Tyler et al., 2015, 2022). Thus, those familiar with TRT theory, the associated treatment principles, and the treatment model for hyperacusis should be at ease with this counseling approach. For those who may not be familiar with these counseling concepts and principles, we plan to offer a counseling package, including a fully scripted protocol and a topical checklist together with a companion set of visual aids and diagrams. We also will include a list of selected publications that provide theoretical and clinical background for the counseling protocol and the protective management and sound treatment strategies that are the foundational components of our transitional intervention (much of which is reviewed by Cherri et al., 2024; Eddins et al., 2024; and Formby et al., 2024, in this issue). The treatment device, offered with and without LS, and the companion fitting protocol and software are technically innovative (see Eddins et al., 2024); together, they represent a comprehensive package that has taken years to plan, develop, refine, and test. Although not trivial to implement, we expect that most audiologists with experience and expertise in fitting hearing aids will readily become facile in implementing the protective management and therapeutic sound treatment component of the intervention. In due course, we expect to offer both the counseling protocol and the protected sound management and treatment device, including the fitting protocol and software, in a cost-effective commercial package for clinical application of the transitional intervention for primary hyperacusis. This same package may also offer a protocol for treating hearing-impaired individuals with reduced auditory dynamic ranges for loudness, with or without decreased sound tolerance. As reported by Formby et al. (2015, 2017), these sound-sensitive individuals may reject amplification. Many of these individuals, however, can become successful hearing-aid users with treatment to broaden their dynamic ranges using similar intervention principles to those described in this report. Controlled trials will be needed in the future to confirm and extend the utility of this and related interventions for hyperacusis.
Data Availability Statement
Selected data sets may be made available to investigators upon reasonable request.
Acknowledgments
This work was supported by National Institute for Deafness and other Communicative Disorders Grant R21DC015054 (C. F. and D. A. E.). Our efforts to develop and evaluate the transitional intervention reported in this field trial were supported by the National Institutes of Health (Public Health Service Award R21DC015054). The authors gratefully acknowledge the voluntary participation of all study participants in this trial; the late Bryan Pollard and Hyperacusis Research Limited for assistance in promoting participant recruitment; Linda Jüris for her aid in translating our TS-H from Swedish to English; and our program officer, Roger Miller, and the National Institute on Deafness and Other Communication Disorders for allowing multiple extensions of the award period without which this trial could not have been completed during the COVID-19 outbreak.
Appendix A
Pure-Tone Thresholds
Audiometric air-conduction thresholds and qualifying LDLs measured in dB HL at the intake visit.
| Participant | Right | Frequency (Hz) | Frequency (Hz) | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 250 | 500 | 1000 | 1500 | 2000 | 3000 | 4000 | 6000 | 8000 | Left | 250 | 500 | 1000 | 1500 | 2000 | 3000 | 4000 | 6000 | 8000 | ||
| Threshold | 0 | 0 | 0 | 0 | 5 | 0 | 5 | 5 | 0 | Threshold | 0 | 0 | 5 | 0 | 5 | 5 | 5 | 0 | 10 | |
| 1 | LDL | 70 | 60 | LDL | 75 | 65 | ||||||||||||||
| Threshold | 5 | 5 | 10 | 10 | 10 | 15 | 10 | 25 | 15 | Threshold | 10 | 5 | 5 | 10 | 10 | 5 | 5 | 10 | 15 | |
| 2 | LDL | 60 | 70 | LDL | 70 | 75 | ||||||||||||||
| Threshold | 5 | 0 | 0 | 5 | 5 | –5 | 0 | 0 | –10 | Threshold | 0 | 0 | 0 | 0 | 5 | 0 | 5 | 5 | 0 | |
| 3 | LDL | 70 | 45 | LDL | 65 | 55 | ||||||||||||||
| Threshold | 5 | 10 | 5 | 20 | 40 | 50 | 50 | 55 | 75 | Threshold | 0 | 5 | 10 | 15 | 30 | 40 | 45 | 55 | 65 | |
| 4 | LDL | 80 | 85 | 105 | LDL | 75 | 75 | 100 | ||||||||||||
| Threshold | 0 | 0 | 5 | 5 | 10 | 10 | 25 | 20 | 50 | Threshold | 0 | 0 | 10 | 10 | 10 | 0 | 20 | 15 | 25 | |
| 5 | LDL | 50 | 65 | LDL | 60 | 60 | ||||||||||||||
| Threshold | 30 | 30 | 25 | 25 | 5 | 0 | 0 | 20 | 15 | Threshold | 30 | 25 | 35 | 20 | 15 | 10 | 5 | 5 | 5 | |
| 6 | LDL | 90 | 65 | 70 | LDL | 95 | 65 | 75 | ||||||||||||
| Threshold | 0 | 0 | 0 | 0 | 5 | 5 | 0 | 0 | 0 | Threshold | 0 | 5 | –5 | 0 | 5 | 10 | 5 | 5 | 10 | |
| 7 | LDL | 60 | 65 | LDL | 55 | 65 | ||||||||||||||
| Threshold | 5 | 5 | 5 | 5 | 10 | 0 | 5 | 10 | 15 | Threshold | 5 | 5 | 5 | 5 | 5 | 10 | 10 | 5 | 20 | |
| 8 | LDL | 75 | 65 | 65 | 75 | LDL | 90 | 70 | 65 | 75 | ||||||||||
| Threshold | 5 | 10 | 15 | 15 | 15 | 20 | 25 | 20 | 20 | Threshold | 10 | 5 | 10 | 15 | 10 | 25 | 25 | 20 | 20 | |
| 9 | LDL | 30 | 30 | LDL | 35 | 45 | ||||||||||||||
| Threshold | 0 | 0 | 0 | 5 | 10 | 5 | 5 | 5 | 10 | Threshold | 5 | 5 | 0 | 5 | 10 | 5 | 10 | 5 | 5 | |
| 10 | LDL | 60 | 55 | LDL | 60 | 55 | ||||||||||||||
| Threshold | 0 | 0 | 0 | 0 | 0 | 15 | 10 | 10 | 15 | Threshold | –5 | 0 | –5 | –5 | –5 | 10 | 15 | 10 | 20 | |
| 11 | LDL | 70 | 70 | LDL | 70 | 65 | ||||||||||||||
| Threshold | 5 | 0 | 0 | 5 | 0 | 5 | 10 | 10 | 5 | Threshold | 0 | –5 | 0 | 0 | 0 | 0 | 0 | 5 | 5 | |
| 12 | LDL | 50 | 35 | LDL | 45 | 40 | ||||||||||||||
Note. LDLs = loudness discomfort levels; dB HL = decibels hearing level.
Appendix B
Tampa Scale for Hyperacusis (TS-H)
Reprinted with permission from Formby and Eddins (2019). Copyright © Craig Formby and David A. Eddins.
1 = strongly disagree
2 = disagree
3 = agree
4 = strongly agree
| 1. I'm afraid that I might injure myself if I am exposed to (loud) sound. | 1 | 2 | 3 | 4 |
| 2. If I were to try to overcome my sensitivity to sound, my pain/discomfort would increase. | 1 | 2 | 3 | 4 |
| 3. My pain/discomfort is signaling that something is seriously wrong with my hearing. | 1 | 2 | 3 | 4 |
| 4. My pain/discomfort would probably be relieved if I were to expose myself to sound. | 1 | 2 | 3 | 4 |
| 5. People aren't taking my condition seriously enough. | 1 | 2 | 3 | 4 |
| 6. My condition has put my hearing at risk for injury for the rest of my life. | 1 | 2 | 3 | 4 |
| 7. Pain/discomfort to sounds always means I have injured my ears. | 1 | 2 | 3 | 4 |
| 8. Just because sounds aggravate my pain/discomfort does not mean they are dangerous. | 1 | 2 | 3 | 4 |
| 9. I'm afraid that I might damage my hearing accidentally with loud sounds. | 1 | 2 | 3 | 4 |
| 10. Simply being careful that I don't put myself in unnecessary sound conditions is the safest thing I can do to prevent my pain/discomfort from worsening. | 1 | 2 | 3 | 4 |
| 11. I wouldn't have this much pain/discomfort if there weren't something potentially dangerous going on in my ears. | 1 | 2 | 3 | 4 |
| 12. Although my condition is painful/uncomfortable, I would be better off if I were exposed to sound. | 1 | 2 | 3 | 4 |
| 13. Pain/discomfort to sound lets me know when to avoid sound so that I don't injure my ears. | 1 | 2 | 3 | 4 |
| 14. It's really not safe for a person with a condition like mine to be exposed to sound. | 1 | 2 | 3 | 4 |
| 15. I can't do all the things normal people do because it's too easy for me to injure my ears in loud sounds. | 1 | 2 | 3 | 4 |
| 16. Even though sound is causing me a lot of pain, I don't think it's actually dangerous. | 1 | 2 | 3 | 4 |
| 17. No one should have to be exposed to sound when he/she is overly sensitive to sound. | 1 | 2 | 3 | 4 |
Appendix C
Hyperacusis Field Trial Exit Survey
The study involved three components:
Counseling regarding the way the auditory system works, the mechanisms of hyperacusis, the benefits of healthy sound and strategies for managing your condition (discussed in detail at the device fitting visit and reviewed during follow up visits)
Sound therapy in the form of a “seashell sound” which was delivered through the devices when using either the open-dome earpieces or the closed earplugs.
Loudness suppression which was active when wearing the closed earplugs with the devices and helped to maintain louder sounds within your range of comfortable listening.
Please describe your experience regarding the Hyperacusis study.
-
(1) Which study component was most beneficial for you?
Why?
-
(2) Which component was least beneficial for you?
Why?
(3) Please describe your overall experience and impressions of study participation.
(4) Did participation and use of the devices affect your quality of life or daily activities in any way? If yes, then please describe.
(5) What specific observations or experiences would you like the research team to know about?
(6) What specific observations or experiences would you share with other individuals considering this treatment?
(7) What do you believe could be done to improve this treatment protocol?
(8) How beneficial was the counseling component of the treatment?
(9) How beneficial was the sound therapy component of the treatment?
(10) How beneficial was the loudness suppression component of the device?
| Not Beneficial | Slightly Beneficial | Moderately Beneficial | Very Beneficial | Extremely Beneficial |
| 1 | 2 | 3 | 4 | 5 |
(11) Please rate the change in your hyperacusis condition post-treatment (time at which you stopped wearing the device) relative to pre-treatment.
| Much Worse | Slightly Worse | No Change | Slightly Better | Much Better |
| 1 | 2 | 3 | 4 | 5 |
Funding Statement
This work was supported by National Institute for Deafness and other Communicative Disorders Grant R21DC015054 (C. F. and D. A. E.). Our efforts to develop and evaluate the transitional intervention reported in this field trial were supported by the National Institutes of Health (Public Health Service Award R21DC015054).
Footnote
Each of the six questionnaires used in this study consists of individual questions requiring categorical responses. The responses to all of the questions are combined into a composite score. The categorical data have properties of an ordinal scale, and the composite score is treated as an interval scale. This common practice for standard questionnaires and surveys can lead to normative data. In interpreting the data collected here, we rely on those norms. The administration, scoring, and analysis of each of the six questionnaires used in this study are based on methods put forth by the original instrument developers. It is important to allow readers to compare the data from these instruments to prior data, which most readers will expect. As a result, we report the data in descriptive and graphical terms in a manner that is consistent with prior literature in the field. There has been no attempt to determine the psychometric properties of the individual questions based on the data reported in this study.
References
- Aazh, H., & Moore, B. C. J. (2017). Factors related to uncomfortable loudness levels for patients seen in a tinnitus and hyperacusis clinic. International Journal of Audiology, 56(10), 793–800. 10.1080/14992027.2017.1335888 [DOI] [PubMed] [Google Scholar]
- Altın, B., Şahin Kamışlı, G. İ., & Aksoy, S. (2023). Validity and reliability study of the Khalfa's Hyperacusis Questionnaire with using ULL in tinnitus patients. European Archives of Oto-Rhino-Laryngology, 280(3), 1485–1492. 10.1007/s00405-022-07727-7 [DOI] [PubMed] [Google Scholar]
- Anari, M., Axelsson, A., Eliasson, A., & Magnusson, L. (1999). Hypersensitivity to sound—Questionnaire data, audiometry and classification. Scandinavian Audiology, 28(4), 219–230. 10.1080/010503999424653 [DOI] [PubMed] [Google Scholar]
- Baguley, D. M., & Andersson, G. (2008). Hyperacusis: Mechanisms, diagnosis, and therapies. Plural. [Google Scholar]
- Beck, A. T., Steer, R. A., & Brown, G. K. (1996). BDI-II, Beck Depression Inventory Manual. Psychological Corporation. [Google Scholar]
- Blaesing, L., & Kroener-Herwig, B. (2012). Self-reported and behavioral sound avoidance in tinnitus and hyperacusis subjects, and association with anxiety ratings. International Journal of Audiology, 51(8), 611–617. 10.3109/14992027.2012.664290 [DOI] [PubMed] [Google Scholar]
- Bläsing, L., Goebel, G., Flötzinger, U., Berthold, A., & Kröner-Herwig, B. (2010). Hypersensitivity to sound in tinnitus patients: An analysis of a construct based on questionnaire and audiological data. International Journal of Audiology, 49(7), 518–526. 10.3109/14992021003724996 [DOI] [PubMed] [Google Scholar]
- Brout, J. J., Edelstein, M., Erfanian, M., Mannino, M., Miller, L. J., Rouw, R., Kumar, S., & Rosenthal, M. Z. (2018). Investigating misophonia: A review of the empirical literature, clinical implications, and a research agenda. Frontiers in Neuroscience, 12, Article 36. 10.3389/fnins.2018.00036 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Carhart, R. (1957). Clinical determination of abnormal auditory adaptation . Archives of Otolaryngology, 65(1), 32–39. 10.1001/archotol.1957.03830190034008 [DOI] [PubMed] [Google Scholar]
- Cherri, D., Formby, C., Secor, C. A., & Eddins, D. A. (2024). Counseling protocol for a transitional intervention for debilitating hyperacusis. Journal of Speech, Language, and Hearing Research. Advance online publication. https://doi.org/10.1044/2023_JSLHR-23-00353 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Routledge. [Google Scholar]
- Cox, R. M., Alexander, G. C., Taylor, I. M., & Gray, G. A. (1997). The Contour Test of loudness perception. Ear and Hearing, 18(5), 388–400. 10.1097/00003446-199710000-00004 [DOI] [PubMed] [Google Scholar]
- Dozier, T. H., Lopez, M., & Pearson, C. (2017). Proposed diagnostic criteria for misophonia: A multisensory conditioned aversive reflex disorder. Frontiers in Psychology, 8, Article 1975. 10.3389/fpsyg.2017.01975 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Eddins, D. A., Armstrong, S., Juneau, R., Hutchison, P., Cherri, D., Secor, C. A., & Formby, C. (2024). Device and fitting protocol for a transitional intervention for debilitating hyperacusis. Journal of Speech, Language, and Hearing Research. Advance online publication. https://doi.org/10.1044/2024_JSLHR-23-00359 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Eddins, D. A., Formby, C., & Armstrong, S. (2020). Method for treating debilitating hyperacusis (US Patent No. 10582286). U.S. Patent and Trademark Office. [Google Scholar]
- Fackrell, K., Fearnley, C., Hoare, D. J., & Sereda, M. (2015). Hyperacusis Questionnaire as a tool for measuring hypersensitivity to sound in a tinnitus research population. BioMedical Research International, 2015, Article e290425. 10.1155/2015/290425 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fackrell, K., & Hoare, D. J. (2018). Scales and questionnaires for decreased sound tolerance. In Fagelson M. & Baguley D. M. (Eds.), Hyperacusis and disorders of sound intolerance: Clinical and research perspectives (pp. 43–58). Plural. [Google Scholar]
- Filion, P. R., & Margolis, R. H. (1992). Comparison of clinical and real-life judgments of loudness discomfort. Journal of the American Academy of Audiology, 3(3), 193–199. [PubMed] [Google Scholar]
- Fioretti, A., Tortorella, F., Masedu, F., Valenti, M., Fusetti, M., & Pavaci, S. (2015). Validity of the Italian version of Khalfa's questionnaire on hyperacusis. Acta Otorhinolaryngologica Italica: Organo Ufficiale Della Societa Italiana Di Otorinolaringologia E Chirurgia Cervico-Facciale, 35(2), 110–115. [PMC free article] [PubMed] [Google Scholar]
- Formby, C., & Eddins, D. A. (2019). The Tampa Scale for Hyperacusis (TS-H): English translation of the Swedish language version of the scale [Unpublished manuscript].
- Formby, C., Gold, S., Keaser, M., Block, K., & Hawley, M. (2007). Secondary benefits from tinnitus retraining therapy: Clinically significant increases in loudness discomfort level and expansion of the auditory dynamic range. Seminars in Hearing, 28(4), 227–260. 10.1055/s-2007-990713 [DOI] [Google Scholar]
- Formby, C., Hawley, M., Sherlock, L., Gold, S., Payne, J., Brooks, R., Parton, J., Juneau, R., Desporte, E., & Siegle, G. (2015). A sound therapy-based intervention to expand the auditory dynamic range for loudness among persons with sensorineural hearing losses: A randomized placebo-controlled clinical trial. Seminars in Hearing, 36(2), 77–110. 10.1055/s-0035-1546958 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Formby, C., Secor, C. A., Cherri, D., & Eddins, D. A. (2024). Background and rationale for a transitional intervention for debilitating hyperacusis. Journal of Speech, Language, and Hearing Research. Advance online publication. 10.1044/2023_JSLHR-23-00352 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Formby, C., Sherlock, L. P., Hawley, M. L., & Gold, S. L. (2017). A sound therapy-based intervention to expand the auditory dynamic range for loudness among persons with sensorineural hearing losses: Case evidence showcasing treatment efficacy. Seminars in Hearing, 38(1), 130–150. 10.1055/s-0037-1598069 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Frisch, M. B., Clark, M. P., Rouse, S. V., Rudd, M. D., Paweleck, J. K., Greenstone, A., & Kopplin, D. A. (2005). Predictive and treatment validity of life satisfaction and the Quality of Life Inventory. Assessment, 12(1), 66–78. 10.1177/1073191104268006 [DOI] [PubMed] [Google Scholar]
- Gold, S., & Formby, C. (2017). Structured counseling for auditory dynamic range expansion. Seminars in Hearing, 38(01), 115–129. 10.1055/s-0037-1598068 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Goldstein, B., & Shulman, A. (1996). Tinnitus—Hyperacusis and the loudness discomfort level test—A preliminary report. The International Tinnitus Journal, 2, 83–89. [PubMed] [Google Scholar]
- Halfaker, D. A., Akeson, S. T., Hathcock, D. R., Mattson, C., & Wunderlich, T. L. (2011). 3—Psychological aspects of pain. In Lennard T. A., Walkowski S., Singla A. K., & Vivian D. G. (Eds.), Pain procedures in clinical practice (3rd ed., pp. 13–22). Hanley & Belfus. 10.1016/B978-1-4160-3779-8.10003-X [DOI] [Google Scholar]
- Hawley, M., Keaser, M., & Formby, C. (2007). Predicting hyperacusis in tinnitus patients. Seminars in Hearing, 28(4), 261–275. 10.1055/s-2007-990714 [DOI] [Google Scholar]
- Hazell, J. W. P., Sheldrake, J. B., & Graham, R. L. (2002). Decreased sound tolerance: Predisposing factors, triggers and outcomes after TRT. In Patuzzi R. (Ed.), Proceedings of the Seventh International Tinnitus Seminar (pp. 255–261). University of Western Australia. https://www.semanticscholar.org/paper/Decreased-sound-tolerance-%3A-predisposing-factors-%2C-Jwpl-Graham/9461f01d5dc65f322d0c1d66cfdc7f5586d1131e [Google Scholar]
- Henry, J. A. (2022). Sound therapy to reduce auditory gain for hyperacusis and tinnitus. American Journal of Audiology, 31(4), 1067–1077. 10.1044/2022_AJA-22-00127 [DOI] [PubMed] [Google Scholar]
- Hirsh, I. J., Davis, H., Silverman, S. R., Reynolds, E. G., Eldert, E. & Benson, R. W. (1952). Development of materials for speech audiometry. Journal of Speech and Hearing Disorders, 17(3), 321–337. 10.1044/jshd.1703.321 [DOI] [PubMed] [Google Scholar]
- Jahn, K. N. (2022). Clinical and investigational tools for monitoring noise-induced hyperacusis. The Journal of the Acoustical Society of America, 152(1), 553–566. 10.1121/10.0012684 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jastreboff, P. J., & Hazell, J. P. W. (2004). Tinnitus retraining therapy: Implementing the neurophysiological model. Cambridge University Press. 10.1017/CBO9780511544989 [DOI] [Google Scholar]
- Jastreboff, P. J., & Jastreboff, M. M. (2000). Tinnitus retraining therapy (TRT) as a method for treatment of tinnitus and hyperacusis patients. Journal of the American Academy of Audiology, 11(3), 162–177. 10.1055/s-0042-1748042 [DOI] [PubMed] [Google Scholar]
- Jastreboff, P. J., & Jastreboff, M. M. (2014). Treatments for decreased sound tolerance (hyperacusis and misophonia). Seminars in Hearing, 35(2), 105–120. 10.1055/s-0034-1372527 [DOI] [Google Scholar]
- Jastreboff, P. J., & Jastreboff, M. M. (2016). Tinnitus and decreased sound tolerance. In Wackym P. A. & Snow J. B. (Eds.), Ballenger's otorhinolaryngology: Head and neck surgery (18th ed., Vol. 1, pp. 391–404). People's Medical Publishing House. [Google Scholar]
- Juneau, R. P., Desporte, E. J., Major, M., Siegle, G., & Tanner, B. (2010). Self-forming in-the-ear hearing aid with conical stent (US Patent No. 7,778,434). US Patent and Trademark Office. [Google Scholar]
- Jüris, L., Andersson, G., Larsen, H. C., & Ekselius, L. (2014). Cognitive behaviour therapy for hyperacusis: A randomized controlled trial. Behaviour Research and Therapy, 54, 30–37. 10.1016/j.brat.2014.01.001 [DOI] [PubMed] [Google Scholar]
- Khalfa, S., Dubal, S., Veuillet, E., Perez-Diaz, F., Jouvent, R., & Collet, L. (2002). Psychometric normalization of a hyperacusis questionnaire. ORL: Journal for Oto-Rhino-Laryngology, Head and Neck Surgery, 64(6), 436–442. 10.1159/000067570 [DOI] [PubMed] [Google Scholar]
- Miller, R. P., Kori, S. H., & Todd, D. D. (1991). The Tampa Scale. The Clinical Journal of Pain, 7(1), Article 51. 10.1097/00002508-199103000-00053 [DOI] [Google Scholar]
- Munro, K. J., & Patel, R. K. (1998). Are clinical measurements of uncomfortable loudness levels a valid indicator of real-world auditory discomfort? British Journal of Audiology, 32(5), 287–293. 10.3109/03005364000000079 [DOI] [PubMed] [Google Scholar]
- Nasreddine, Z. S., Phillips, N. A., Bédirian, V., Charbonneau, S., Whitehead, V., Collin, I., Cummings, J. L., & Chertkow, H. (2005). The Montreal Cognitive Assessment, MoCA: A brief screening tool for mild cognitive impairment. Journal of the American Geriatrics Society, 53(4), 695–699. 10.1111/j.1532-5415.2005.53221.x [DOI] [PubMed] [Google Scholar]
- Nelting, M., & Finlayson, N. (2004). GÜF -Gerauschuberempfindlichkeits-Fragebogen. Hogrefe. [Google Scholar]
- Nelting, M., Rienhoff, N. K., Hesse, G., & Lamparter, U. (2002). The assessment of subjective distress related to hyperacusis with a self-rating questionnaire on hypersensitivity to sound. Laryngo-Rhino-Otologie, 81(05), 327–334. 10.1055/s-2002-28342 [DOI] [PubMed] [Google Scholar]
- Nilsson, M., Soli, S. D., & Sullivan, J. A. (1994). Development of the Hearing in Noise Test for the measurement of speech reception thresholds in quiet and in noise. The Journal of the Acoustical Society of America, 95(2), 1085–1099. 10.1121/1.408469 [DOI] [PubMed] [Google Scholar]
- Noreña, A. J., & Chery-Croze, S. (2007). Enriched acoustic environment rescales auditory sensitivity. NeuroReport, 18(12), 1251–1255. 10.1097/WNR.0b013e3282202c35 [DOI] [PubMed] [Google Scholar]
- Punch, J., Joseph, A., & Rakerd, B. (2004). Most comfortable and uncomfortable loudness levels: Six decades of research. American Journal of Audiology, 13(2), 144–157. 10.1044/1059-0889(2004/019) [DOI] [PubMed] [Google Scholar]
- Sammeth, C. A., Preves, D. A., & Brandy, W. T. (2000). Hyperacusis: Case studies and evaluation of electronic loudness suppression devices as a treatment approach. Scandinavian Audiology, 29(1), 28–36. 10.1080/010503900424570 [DOI] [PubMed] [Google Scholar]
- Sawilowsky, S. (2009). New effect size rules of thumb. Journal of Modern Applied Statistical Methods, 8(2), 597–599. 10.22237/jmasm/1257035100 [DOI] [Google Scholar]
- Scherer, R. W., & Formby, C. (2019). Effect of tinnitus retraining therapy vs standard of care on tinnitus-related quality of life: A randomized clinical trial. JAMA Otolaryngology–Head & Neck Surgery, 145(7), 597–608. 10.1001/jamaoto.2019.0821 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Searchfield, G. D., & Selvaratnam, C. (2018). Hearing aids for decreased sound tolerance and minimal hearing loss: Gain without pain. In Fagelson M. & Baguley D. M. (Eds.), Hyperacusis and disorders of sound intolerance: Clinical and research perspectives (1st ed., pp. 223–239). Plural. [Google Scholar]
- Sheldrake, J., Diehl, P. U., & Schaette, R. (2015). Audiometric characteristics of hyperacusis patients. Frontiers in Neurology, 6, Article 105. 10.3389/fneur.2015.00105 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sherlock, L. P., & Formby, C. (2005). Estimates of loudness, loudness discomfort, and the auditory dynamic range: Normative estimates, comparison of procedures, and test-retest reliability. Journal of the American Academy of Audiology, 16(2), 85–100. 10.3766/jaaa.16.2.4 [DOI] [PubMed] [Google Scholar]
- Tillman, T. W., & Carhart, R. (1966). An expanded test for speech discrimination utilizing CNC monosyllabic words: Northwestern University Auditory Test No. 6. USAF School of Aerospace Medicine. [DOI] [PubMed] [Google Scholar]
- Tyler, R. S., Noble, W., Coelho, C., Roncancio, E., & Jun, H. J. (2015). Tinnitus and hyperacusis. In J. Katz, M. Chasin, K. English, L. Hood, & K. Tillery (Eds.), Handbook of clinical audiology (7th ed., pp. 647–658). Wolters Kluwer Health. [Google Scholar]
- Tyler, R., Perreau, A., & Mancini, P. (2022). Hyperacusis. In Tyler R. S. & Perreau A. (Eds.), Tinnitus treatment: Clinical protocols (2nd ed., pp. 165–197). Thieme Publishers. 10.1055/b-0042-189044 [DOI] [Google Scholar]
- Tyler, R., Perreau, A., Mohr, A. M., Ji, H., & Mancini, P. C. (2020). An exploratory step toward measuring the “meaning of life” in patients with tinnitus and in cochlear implant users. Journal of the American Academy of Audiology, 31(4), 277–285. 10.3766/jaaa.19022 [DOI] [PubMed] [Google Scholar]
- Tyler, R. S., Pienkowski, M., Roncancio, E. R., Jun, H. J., Brozoski, T., Dauman, N., Coelho, C. B., Andersson, G., Keiner, A. J., Cacace, A. T., Martin, N., & Moore, B. C. J. (2014). A review of hyperacusis and future directions: Part I. Definitions and manifestations. American Journal of Audiology, 23(4), 402–419. 10.1044/2014_AJA-14-0010 [DOI] [PubMed] [Google Scholar]
- Zaugg, T. L., Thielman, E. J., Griest, S., & Henry, J. A. (2016). Subjective reports of trouble tolerating sound in daily life versus loudness discomfort levels. American Journal of Audiology, 25(4), 359–363. 10.1044/2016_AJA-15-0034 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Selected data sets may be made available to investigators upon reasonable request.