Patients with Tinnitus: Their Perspective on Sound Generators

Abstract

Objective:

Sound therapy (ST) is a well-known treatment option for tinnitus patients. However, patient perspectives on this option remain scarce. This study investigated the perspectives of first-time sound generator users on ST for chronic tinnitus, focusing on their perceptions of the sound generators and the effects on tinnitus.

Methods:

The subjects were 29 adult patients with chronic tinnitus who received ST using sound generators for the first time. In the 3-week ST trial, In the 3-week ST trial, hearing aids with built-in sound generators built idelivered broadband noise at a low and individualized sound level. Semistructured interviews were conducted 1 month after the ST trial. Using an interpretive description approach, themes were captured qualitatively to describe patients’ subjective experience of the ST.

Results:

After performing inductive coding on the dataset, five main themes emerged: (1) effects while wearing the generators; (2) effects after removing the generators; (3) perception of the generated noise; (4) untargeted effects; and (5) purchasing sound generators. Each theme was further subdivided into one to three subthemes.

Conclusions:

The study highlighted uncertainties regarding the overall perceived benefits of short-term ST, with results influenced by initial noise hypersensitivity, hearing loss, and noise appreciation. Future research should examine the benefits of ST after controlling for these variables and compare the effects of different types of tailored sound on each dimension of tinnitus. Understanding how noise physiologically modulates tinnitus both during and after exposure is crucial to better counsel patients on what to expect.

KEY MESSAGES

• (1)First-time users of sound therapy for tinnitus described their experience regarding the use of sound generators for 3 weeks.
• (2)Five themes emerged, pertaining to the effects while wearing or after removing sound generators, perception of the noise, purchasing generators, and other effects.
• (3)Our study informs both clinicians and patients about what to expect when choosing sound therapy for tinnitus.

INTRODUCTION

Tinnitus, an auditory sensation arising without any external source, affects up to 30% of the general population.[1]Sensorineural hearing loss (HL) is considered the primary predictor of tinnitus and may play a role in tinnitus development through neural overcompensation and hyperexcitability.[2,3] In this perspective, increasing auditory input could recalibrate auditory gain, enhance neural plasticity, and provide distraction by enhancing background noise.[4,5] Sound therapy (ST), which can involve amplification devices, sound generators, or both, is indeed the preferred mode of audiological tinnitus management in many countries.[6] Nowadays, most hearing aids have built-in sound generators that deliver a controlled, constant sound in addition to providing amplification. Channel-by-channel adjustment in these devices allows for the programming of various bandpass noises, ranging from broadband “white” noise to a tinnitus-centered “customized” noise. Despite the potential to alleviate social isolation, loneliness, depression, and to improve quality of life,[7] hearing aids have consistently shown low uptake among users. First, only half of individuals with significant HL who seek assistance actually acquire hearing aids.[8] Furthermore, among hearing aid owners, up to a quarter never use them.[9] So, why do patients choose to leave these beneficial devices unused? Qualitative studies have identified several reasons for nonuse, including the nonacceptance of HL,[8] perception of only a mild impairment, disappointment with the benefits of hearing aids,[10] stigma associated with aging and reduced intelligence,[11] auditory fatigue and discomfort,[12] difficulty changing batteries,[13] high costs, and insufficient follow-up by practitioners.[14] Therefore, gaining insight into patients’ perspectives and addressing these barriers is crucial for providing effective patient care and achieving successful treatment outcomes.

In a recent qualitative study aimed at understanding how individuals with tinnitus experienced and benefitted from their sound-generating devices, patients described a sense of escapism and control as positive aspects of using sound generators.[15] Hearing aids represented forms of social and medical validation, as they helped the patient’s entourage and the medical community recognize tinnitus as a clinical condition. These positive findings, however, reflect information collected from 10 participants out of the 72 invited to participate in the study. It is likely that only those who experienced benefits volunteered to share their perception, especially since the authors acknowledged that their study addressed the process that facilitates tinnitus reduction. In addition, it is unknown whether participants were first-time users and how much time they spent each day wearing their devices. For instance, listening to sound for 8 hours a day while already dealing with tinnitus might be challenging, especially considering the additional stigma and limitations associated with wearing hearing aids.[12] Overall, patients’ perspectives regarding the satisfaction with sound generators remain largely unaddressed.

The present study is the qualitative counterpart to our previous paper, which reported the effect of sound generators on the synchronous psychophysical and psychological modulation of hyperacusis and tinnitus loudness in first-time users.[16] Overall findings regarding tinnitus loudness were mixed, with approximately half of the participants experiencing a decrease and the other half experiencing an increase in their perception of loudness after 3 weeks of acoustic stimulation. To examine participants’ perspectives on the effects of sound generators on their tinnitus, we conducted semistructured interviews. Participants’ experiences with tinnitus, more generally, were published in a previous paper.[17] Herein, we report qualitative analyses specifically focused on the ST. Three main aspects were explored: (1) the perceived effects of sound generators on tinnitus; (2) the appreciation of the noise produced by the generators; and (3) the motivation to buy sound generators after the trial. We followed the Consolidated Criteria for Reporting Qualitative Research (COREQ)[18] to make sure that the study responded to the highest requirements of a qualitative study.

MATERIALS AND METHODS

Participants

Twenty-nine adults (22 men and 7 women) with a mean age of 55.0 ± 12.0 years participated in a proof-of-concept study that involved wearing sound generators for 3 weeks.[16] Participants were recruited through diverse channels, including the lab’s database, tinnitus association websites and Facebook posts, local journals, and audiology clinic advertisements in Montreal, Canada. Inclusion criteria were self-reported tinnitus for more than 6 months (mean = 11.6 ± 10.0 years, range = 1–40 years), receiving ST for the first time, and normal hearing (NH) or NH sloping to mild-to-severe HL. Exclusion criteria included wearing hearing aids, having conductive HL, neurological disorders, uncontrolled medical conditions (e.g., hypertension, diabetes), psychological or psychiatric treatment, and medications affecting the nervous system.

Participants were allocated either to the normal/near-NH group or to the HL group, consistent with our previous study.[16] Participants with audiometric thresholds ≤40 dB HL at all standard frequencies (0.25–8 kHz) were classified into the normal to mild HL group (NH), n = 16), while the remaining participants were classified into the moderate to severe HL group (n = 13). The NH participants were, on average, younger than the HL participants (mean age 48.1 ± 9.9 and 63.5 ± 8.3 years, respectively, using Student’s t-test, t (27) = 4.55, P < 0.001). Tinnitus duration did not differ between groups (13.6 ± 12.4 and 9.3 ± 6.1 years, respectively, P = 0.27).

Procedure

Fitting of the Sound Generators

For the ST part of the study, ST was provided by Pure Life 7mi Siemens open-fitted Behind-the-ear (BTE) hearing aids, with the sound generator option turned on and the amplification option turned off. The chosen noise was broadband and not centered around the tinnitus frequency, as there is no consistent or widely accepted method for tinnitus matching.[19] The stimulation level was first obtained with the probe placed near the eardrum and the sound generator fitted in the ear but turned off, while the patient was seated in a sound booth. Each hearing aid was connected to the Siemens Connexx 7 system, which allowed the experimenter to manually modify the noise level provided in each frequency band separately. The sound was first set approximately 40 dB at the eardrum level for frequencies ranging from 2 to 10 kHz for each ear separately. Both sound generators were then turned on simultaneously, and the participant was asked whether the sound level was comfortable enough for daily use. If the sound was judged to be uncomfortable, the level was slightly reduced until an acceptable level was achieved. The noise level gain was obtained by subtracting the noise level measured with the hearing aid turned off from the level measured with the hearing aid turned on, both measured with a probe at the eardrum.

During the trial, participants—who were all first-time users—were encouraged to wear their sound generators for at least 8 hours per day. Data logging from the sound generators was obtained for 16 participants and showed an average usage of 9 h/day (4.5–15 h/day). No specific tinnitus counseling, such as the one provided in Tinnitus Retraining Therapy, was given. Advice was given on the use and maintenance of the equipment. One month after the trial ended, participants were requested to provide feedback on their experience with the sound generators through semistructured, face-to-face interviews. The interviews were conducted in French by five research assistants, all of whom were audiology or psychology students in the laboratory, and were audiorecorded. On average, the interviews lasted 13 minutes (range: 5–42 minutes). The interviews analyzed in the present study addressed patients’ perspectives about using sound generators for the first time and their effects on tinnitus. Participants’ considerations regarding the purchase of sound generators after wearing them for 3 weeks were also of interest. Clarification and follow-up questions were asked when necessary. The questions are outlined in Table 1.

Table 1.

Structure of the Interviews with Main, Additional, and Clarification Questions.

Main and Additional Questions	Clarification Questions
1. Have you noticed any change in your tinnitus during your participation in this study?	Can you expand on your experience?
If yes, since when?
Did the level of your tinnitus vary during the study?Did your tinnitus bother you during the study?
2. Can you tell me about your experience and perception related to the sound generators, those on your ears?	Can you give me some examples? Can you tell me a little more?
How did you find the use of these sound generators?
How did you perceive the noise generated by the sound generators?
3. Would you consider spending money to buy sound generators for your tinnitus?
If yes, why?
If not, why?

Qualitative Analysis

The conceptual framework for the analysis was based on interpretive description,[20] which is an inductive analytic method aimed to document and understand clinical phenomena, i.e., the individual perception of sound generators on tinnitus during the trial. Perception of sound generators was explored here in first-time users without the potential bias of including only “satisfied participants”. The question was: “what is it like to wear sound generators when having tinnitus?”. The aim of this approach is to identify themes and patterns in subjective perceptions and to create an interpretive description that enhances clinical understanding.

The dataset consisted of text transcribed verbatim from recorded interviews. Inductive coding was performed by authors B.V. and C.B. using the qualitative data analysis software package QDA Miner 6.0.2 (Provalis Research, Montréal, QC, Canada). The first step was to perform a thematic analysis[21] to identify the main themes derived from the interviews. The researchers independently identified the codes that most accurately represented the ideas expressed by the participants. Each code was then labeled as positive or negative, depending on the context provided by the participants. After a consensus was reached among researchers, the codes were categorized into multiple themes and subthemes. Some themes directly aligned with the wording of the questions, while others were derived from the interview responses. Some codes, particularly those associated with hearing aids, were not the primary focus of this study and have been previously explored in a separate qualitative study.[12] We include in then Supplementary material the codes that were deemed irrelevant to the specific research questions. Coding was iterative, and all quotes were re-analyzed a second time to ensure that they agreed with the themes already identified. Finally, inter-rater agreement was reached by cross-checking at least 50% of the interviews.

The second step consisted of a frequency analysis. The number of interviews in which a theme was cited was calculated; hence, a theme was more popular when cited in more interviews (i.e., by more participants). Note that a theme could be cited twice if it was mentioned both positively and negatively in the same statement. All interviews were analyzed in the original language (French). For publication purposes, selected quotes were freely translated from French to English by the two authors who transcribed and analyzed the text. Statements are reported within quotation marks, followed by the participant number and hearing status. For instance, “NH04” means participant number 4 in the NH group. The distinction between NH and hearing-impaired participants is important, as the behavioral and experimental parts of the project revealed very significant effect of the sound generators on these two populations.[16]

RESULTS

Relevant sociodemographic and tinnitus characteristics, as well as pretreatment and 1-month post-treatment scores for tinnitus and hyperacusis are presented in Table 2.

Table 2.

Sociodemographic, Tinnitus Characteristics and Questionnaire Scores Before and After ST (from Reference 16) for the 29 Participants Interviewed.

Participants	Hearing Group	Sex	Age (Years)	Tinnitus Duration (Years)	Tinnitus Side	Tinnitus Etiology	HQpre	HQpost	THQpre	THQpost
NH04	NH	F	46	10	Bilateral	Idiopathic	10	9	390	345
NH05	NH	F	41	NA	Left	Idiopathic	16	16	370	800
NH09	NH	M	43	NA	Bilateral	Noise exposure	22	24	1345	1180
NH14	NH	F	42	4	Bilateral	Idiopathic	34	33	1175	1540
NH15	NH	M	55	2	Bilateral	Idiopathic	9	4	816	375
NH17	NH	M	51	12	Bilateral	Idiopathic	14	16	1080	1030
NH21	NH	F	59	1	Bilateral	TMD	12	9	650	270
NH25	NH	M	61	9	Bilateral	Idiopathic	11	6	250	305
NH31	NH	M	59	12	Bilateral	Repetitive OM	8	6	530	890
NH35	NH	M	58	15	Bilateral	Idiopathic	18	29	1410	1170
NH39	NH	F	42	10	Bilateral	Idiopathic	16	15	300	190
NH41	NH	M	60	40	Bilateral	Idiopathic	15	9	265	325
NH43	NH	M	44	1	Bilateral	Idiopathic	16	20	720	710
NH45	NH	M	29	10	Bilateral	Idiopathic	32	26	674	1008
NH46	NH	F	46	32	Bilateral	Idiopathic	22	20	730	695
NH47	NH	M	33	32	Bilateral	Idiopathic	13	7	770	270
HL01	HL	M	48	NA	Bilateral	Idiopathic	18	22	1690	1715
HL02	HL	M	66	15	Bilateral	Idiopathic	10	6	560	470
HL03	HL	M	57	2	Bilateral	Idiopathic	29	26	1625	1650
HL06	HL	M	51	5	Bilateral	Idiopathic	21	32	435	300
HL08	HL	M	65	1	Bilateral	Idiopathic	24	29	1280	1180
HL13	HL	M	61	15	Bilateral	Idiopathic	12	23	520	835
HL22	HL	M	77	15	Bilateral	Idiopathic	27	26	1260	900
HL24	HL	F	59	3	Bilateral	Idiopathic	27	29	1735	2060
HL26	HL	M	66	10	Bilateral	Noise exposure	33	33	1140	1140
HL28	HL	M	75	6	Bilateral	Idiopathic	19	20	500	510
HL29	HL	M	69	10	Bilateral	Noise exposure	19	16	442	575
HL36	HL	M	65	20	Bilateral	Idiopathic	19	10	850	520
HL38	HL	M	66	10	Bilateral	Head trauma	7	13	915	710

F = female, HL = hearing loss, HQ = Hyperacusis Questionnaire (possible scores from 0 to 42), M = male, N/A = >6 months, exact duration unknown, NH = normal hearing, OM = otitis media, pre = before treatment, post = 1-month follow-up, THQ = Tinnitus Handicap Questionnaire (possible scores from 0 to 100), TMD = temporomandibular disorder

Five main themes were retrieved from the thematic analysis: (1) effects while wearing generators, (2) effects after removing generators, (3) perception of the generated sound, (4) untargeted effects, and (5) purchasing sound generators. The themes were divided into subthemes to provide a more specific description of the participants’ perspectives and were labeled as positive or negative based on the contextual information. All themes, subthemes, and features are presented in Table 3. Participants are identified, as some mentioned both positive and negative aspects.

Table 3.

Themes and Subthemes Retrieved from the Thematic Analysis on the Perception of Sound Generators.

Themes and Subthemes	Positive Aspects	Negative Aspects
1. Effects while wearing generators	N = 17	N = 28
Tinnitus loudness	Softer or masked (n = 14) NH: #05, #09, #14, #15, #17, #35, #43, #46 HL: #06, #08, #22, #26 #28, #29	Louder or unchanged (n = 15) NH: #04, #05, #09, #15, #25, #35, #46, #47 HL: #01, #03, #06, #08, #24, #29, #38
Tinnitus annoyance	Less annoying, less stressful (n = 11) NH: #05, #09, #14, #15, #17, #35, #45 HL: #06, #08, #22, #26	More annoying or unchanged (n = 17) NH: #17, #21, #31, #39, #41, #43, #46, #47 HL: #01, #02, #03, #04, #08, #13, #24, #28, #38
Tinnitus awareness	Reduced awareness, easier to forget (n = 8) NH: #09, #21, #35, #41 HL: #06, #08, #22, #28	Attention-grabbing (n = 5) NH: #14, #45 HL: #02, #13, #36
Hope and empowerment over tinnitus	Concrete action, feeling non-powerless, illusion of remission, expectation of remission (n = 3) NH: #14, #15, #17
2. Effects after removing generators	N = 7	N = 18
During trial period	Softer or inhibited (n = 6) NH: #05, #09, #17 HL: #22, #28, #38	Louder or more annoying (n = 9) NH: #05, #21, #35, #43, #47 HL: #08, #24, #26, #28 Difficulties to fall asleep (n = 1) NH: #43
After trial period	Less perceived (n = 1) NH: #14	Increased awareness, louder, or unchanged (n = 11) NH: #05, #17, #31, #39 HL: #02, #03, #06, #08, #22, #28, #29
3. Perception of the generated sound	N = 18	N = 15
General perception	Easy to habituate to, less perceived over time Comfortable, neutral, relaxing, balanced both ears (n = 18) NH: #04, #09, #17, #21, #35, #41, #43, #46, #47 HL: #01, #06, #08, #13, #22, #24, #26, #28, #29	Difficult to habituate to, unpleasant, exhausting, disturbing (n = 13) NH: #05, #14, #15, #21, #25, #31, #35, #39, #41, #43, #46 HL: #03, #38
Comparison with tinnitus	More pleasant than tinnitus (n = 3) NH: #35 HL: #22, #28	Cumulative effect of noise and tinnitus, too different or too similar to tinnitus, less pleasant than tinnitus (n = 8) NH: #05, #21, #31, #41 HL: #03, #08, #24, #38
4. Untargeted effects	N = 8	N = 10
Effects on hyperacusis	Improved hyperacusis (n = 3) NH: #09, #14, #45
Effects on hearing	Did not interfere with hearing (n = 2) NH: #14, #46	Reduced speech intelligibility, muffled effect Decreased sound quality (n = 8) NH: #04, #05, #39, #43 HL: #03, #06, #26, #36
Other effects	Improved stress, relaxation, and concentration (n = 7) NH: #09, #14, #35, #45 HL: #06, #08, #22	Headaches and balance issues (n = 2) NH: #47, #31
5. Purchasing generators	N = 3	N = 28
Personal factors	Improved hyperacusis, stress, concentration (n = 2) NH: #09 HL: #08 Expecting a lasting effect (n = 1) HL: #22	Lack of effectiveness (n = 17) NH: #04, #05, #14, #15, #17, #25, #31, #41, #46, #47 HL: #01, #13, #22, #24, #29, #36, #38 Tinnitus not severe enough (n = 7) NH: #35, #39, #43, #45 HL: #08, #26, #36 Disliked noise (n = 6) NH: #05, #14, #15, #35 HL: #03, #38 Expected silence (n = 1) NH: #21
External factors		Too expensive (n = 8) NH: #05, #17, #35 HL: #02, #06, #22, #26, #38 Physical discomfort (n = 2) HL: #26, #38 Unsightly or social barrier (n = 2) HL: #22, #28

Participants who reported both positive and negative aspects are given in bold. HL = hearing loss, N = the number of participants and can overlap between positive and negative aspects, and between themes and subthemes, NH = normal hearing or near-normal hearing participant

Table 4.

Supplementary Table 1 Additional codes not included in the qualitative analysis.

Codes	Examples
Social acceptance/aesthetics	Discreet
	Social barrier
Physical sensation	Comfortable
	Foreign body
Use of the devices	Easy to manipulate
	Unsuited to daily routine (e.g., sports, water, glasses)
Areas for improvement	Waterproof
	Adjust volume to environment
	Fit to glasses

Theme 1—Effects While Wearing Generators (n = 29)

When participants were asked “Have you noticed any change in your tinnitus during your participation in this study?” the reported changes were gathered into the first theme “Effects while wearing generators,” which represented the “direct” effect of noise on tinnitus, while participants used the sound generators. Multidimensional aspects of tinnitus perception were noted in the responses. Therefore, these aspects were divided into four subthemes: (1) tinnitus loudness that refers to the inherent characteristics of tinnitus, (2) tinnitus annoyance that refers to the psychological/emotional perception of tinnitus, (3) tinnitus awareness that refers to attentional resources directed toward or away from tinnitus, regardless of how loud or annoying it was, and (4) a sense of hope and empowerment over tinnitus. Additional details can be found below.

Tinnitus Loudness (n = 21)

Collectively, 21 participants reported changes in tinnitus loudness, with mixed perceptions observed, both between and within participants. Fourteen participants (8 NH, 6 HL) perceived their tinnitus as softer or masked while wearing the sound generators. In contrast, 15 participants (8 NH, 7 HL) perceived their tinnitus louder than before or did not report any change. Eight participants (5 NH, 3 HL) reported mixed feelings with tinnitus being softer and louder at different time points during the ST trial. For instance, one participant said “Because, in a way, the devices, they managed to cover the tinnitus, but when there was an attack, well it increased again. You know, I… I noticed that I had the… it was the same phenomenon that was occurring there. When I talk, let’s say about… a given moment of… a little tinnitus attack, as I call it, um… it comes back up all of a sudden and then it, it starts ringing louder, um… I had that too when wearing the devices.” (HL08).

Notably, for seven participants, tinnitus “overcame the noise” at some point, meaning that tinnitus grew progressively louder than the noise throughout the day. Regardless of hearing status, the number of participants describing positive effects (n = 14, 8 NH, 6 HL) was roughly comparable to those reporting negative effects (n = 15, 8 NH, 7 HL), a reflection of the unpredictable and variable effects of noise exposure on tinnitus loudness.

Tinnitus Annoyance (n = 26)

Similar to tinnitus loudness, the effects on tinnitus annoyance were both positive (n = 11, 7 NH, 4 HL) and negative (n = 17, 8 NH, 9 HL), and were collectively cited by 26 participants. Seventeen participants found their tinnitus to be equally or even more annoying than before. However, tinnitus was considered less bothersome or less stressful for 11 other participants, as expressed by the following statement: “I started to appreciate the generators, I found that it was good, that it was a less tiring sound [than tinnitus], less stressful… (Did your tinnitus bother you less when you wore the sound generators compared to now when you no longer wear them?) Oh yes, definitely. With the generators, it didn’t happen often that tinnitus passed through the noise, becoming louder. But sometimes they were almost equal.” (NH35). Overall, there did not seem to be any difference between NH and HL participants.

Tinnitus Awareness (n = 13)

Generators played a role in reducing time of awareness of tinnitus, as noted by eight participants (4 NH, 4 HL). The noise helped individuals to disregard tinnitus, making it easier to “forget,” as for one participant, generators “make one forget the noise of tinnitus a little bit” (HL28). Another participant said that generators were effective in diverting attention from the intrusive nature of tinnitus, stating that they were “effective, just for the fact of no longer having to listen to tinnitus, which is often (…) very attention-grabbing, so white noise is definitely something soothing.” (HL06).

In contrast, five participants (2 NH, 3 HL) noted heightened awareness of their tinnitus, stating that the generators “made [them] focus a little more on tinnitus,” notably due to the type of sound generated. They stated that the tinnitus sound changed which resulted in an increased focus, as one participated stated: “[Wearing] the sound generators with a steady, constant sound in my ear, it seems like my focus was on tinnitus, on the tinnitus projected by the generators, and not my own.” (HL13). The novelty of the noise seemed to conflict with the tinnitus sound to which they were habituated. In other words, the sound generated by the devices interfered with their usual coping strategy and prevented the habituation process, as a participant noted “it’s a new sound that is there, all the time, whereas for my tinnitus, a habit was created.” (HL13). In addition, handling and wearing an external device in the ear may have contributed to the awareness of tinnitus, as one person explained “I feel like taking care of the generators, thinking about putting them on, taking them off… that attracts our attention to tinnitus, and, for me, it accentuated the… I felt like it made the feeling of… worsening tinnitus (laughs). Before, I didn’t pay much attention to it, but with the devices, you pay more attention to it and… I feel like I heard them louder.” (NH45).

Hope and Empowerment over Tinnitus (n = 3)

Sound generators also seemed to reinstate a sense of control and influence over tinnitus. Two participants (2 NH) recognized that their internal sounds would not simply vanish and tended to persist despite interventions, and thus viewed sound generators as offering a means to exert some control over their tinnitus. One participant described generators as enabling them to “do something concrete about their tinnitus” and feeling that they were not “just powerless with nothing happening” (NH15). Another participant stated, “I had the illusion for a while that… because there were moments when [my tinnitus] stopped, that maybe it could stop completely, but I think that’s not a very reasonable expectation.” (NH17). The terms “concrete,” “non-powerless,” “illusion,” and “expectation” collectively evoked the concept of individuals experiencing a temporary sense of empowerment and hope for remission, driven by the belief that they could potentially revert to their original state without tinnitus.

Theme 2—Effects after Removing Generators (n = 20)

In addition to the effects experienced while wearing generators, they induced several effects that lasted after the noise offset or after the trial ended. These effects, despite being very distinct—such as tinnitus inhibition versus exacerbation—varied not only in type but also in timelines. During the trial period (n = 14), effects lasted minutes to hours, while after the trial period (n = 12), effects could persist for days to weeks. Regardless of the duration or direction of the effects, these changes were reported by more than two-thirds of the sample (20 participants). Importantly, this was not part of the interview questions which means that participants spontaneously described these changes without being asked. Overall, 18 participants (9 NH, 9 HL) reported a worsening of their tinnitus, while less than half (n = 7, 4 NH, 3 HL) reported an improvement.

During Trial Period (n = 14)

In six participants (3 NH, 3 HL), tinnitus was softer or inhibited at noise offset, at the time of removing the generators. This phenomenon is usually observed in the clinic when assessing residual inhibition and lasts no longer than 1 minute. Here, participants noted that this short-term suppression could last for a “few minutes,” as stated: “there were rare moments when, after removing the sound generators, I had the impression, for a brief period, that I no longer had any tinnitus at all. (Interviewer: How long did that last?) A few minutes.” (NH17). On a slightly longer scale, positive prolonged effects lasting up to hours were also reported “…when I took off the generators, after a few hours, [tinnitus] would come back” (HL28), or “…when I took them off, I was good for a little while” (NH09) or “as soon as I stopped wearing the white noise generator, I wouldn’t say on the first day, but on the following two or three days, I could hardly perceive my tinnitus anymore.” (HL22).

In contrast, nine (5 NH, 4 HL) perceived tinnitus as louder or more annoying, even though it remains unclear whether the worsening was due to an actual increase in loudness, or a heightened focus on tinnitus after continuous masking “Immediately after the generators were removed, everything came back. (…) it’s like water in the face; it comes back to you right away.” (NH35). As tinnitus ended up being louder or more annoying, one participant also reported difficulties falling asleep “removing the generators during the day to sleep, it was perhaps the most difficult phase in the study” (NH43).

After Trial Period (n = 12)

The “after trial period” refers to the period when participants no longer wore the devices, starting at the end of the 3-week trial and ending a month later during the data collection. This subtheme highlights the potential of sound generators to induce a lasting effect on tinnitus. Among our samples, only one participant (NH14) experienced a prolonged reduction in tinnitus awareness, after discontinuing the use of the generators: “I really feel like I have significantly less tinnitus on the left side. I feel that there are certain sounds of my tinnitus that I no longer hear; it seems lighter, as if a layer has been lifted, or… I’m not quite sure how to describe it.” (NH14).

However, among those who noted some improvement after removal of the generators, the effect was not lasting, but rather, for many, their tinnitus tended to worsen. Eleven participants (4 NH, 7 HL) reported that their tinnitus became louder or more noticeable once the trial ended. Notably, permanently removing the generators could lead to a worsening of tinnitus, even when participants perceived an improvement during the trial “It was a little less annoying [during the trial], but since I no longer wear the device, it seems like it’s constant and it’s there. I don’t know if it will ever go away.” (HL29). During the trial, it was not clear whether this was a direct result of noise stimulation or a contrast effect that occurred when tinnitus was no longer masked.

Theme 3—Perception of the Generated Sound (n = 26)

There was no consensus across the sample regarding the perception of the noise, and participants expressed quite contradictory views with respect to their appreciation of the noise, with overall, 18 citing elements in favor of it (9 NH, 9 HL) and 15 against (11 NH, 4 HL). The perceptions were separated into two subthemes, one related to general perception and the other to a comparison with tinnitus.

General Perception (n = 26)

One of the main reasons for disliking the generated sound was that it was very loud: “I found the noise of the generators very loud… If someone suggested that I wear them there, it would have to be less loud (laughs)” (NH14), difficult to get used to: “People here told me that probably we’ll get used to the noise in the second week. Not me.” (NH15). Other reasons included perceiving the noise as unpleasant, disturbing, or even exhausting: “It was the noise that I found tiring (…) I was happy to take it off.” (HL38). Overall, this concerned 13 participants, especially those with NH (11 NH, 2 HL). On the positive side, 18 individuals (9 NH, 9 HL) indicated that they either acclimated to the noise or perceived it as barely noticeable, but it is not clear whether this was a result of environmental masking or habituation. While the noise may have been perceived as annoying initially, participants eventually became accustomed after a certain period: “No, I can’t say it was a bothersome noise once you get used to it.” (NH21) or “The sound generators themselves… it’s still, it’s still bothersome, but you get used to it. (…) I think after the first week, you get used to it.” (NH41). The noise was also found to be comfortable, neutral, or relaxing. Two participants even wore the devices while sleeping. One also appreciated having noise in both ears as it balanced out their asymmetrical tinnitus (HL22).

Comparison with Tinnitus (n = 11)

Noise was often compared to tinnitus (n = 11), i.e., how they matched or mixed with each other (n = 3, 1 NH, 2 HL), or how they conflicted (n = 8, 4 NH, 4 HL). On one side, noise was reported as “less pleasant than tinnitus”: “I prefer my tinnitus to the noise of the device.” (HL03). A cumulative effect of noise and tinnitus was reported (four times), with the generator being described as “just one more sound” preventing from habituation to either tinnitus or noise, as stated by these participants: “[my tinnitus] is composed of two completely different sounds, plus the ‘shhhhh’ of the masker. It’s like having an orchestra in your head, like a symphony. You can’t wear that all day long, seven days a week while hearing three noises. (…) the three together, it’s maddening. Forget it, you can’t. I tested it, some days I put them on, some days I didn’t… I thought it would make me feel well, but in the end, it didn’t help me that much.” (HL24) and “I found that having another sound was like… I mean, I already had one (laughs). The idea is that I didn’t wish for another one… it’s kind of introducing another noise… I wished I didn’t have any noise at all.” (NH21). On the other side, noise was, for three participants (1 NH, 2 HL), perceived as neutral or more pleasant than tinnitus “My tinnitus is very high-pitched, whereas the noise from the generator is lower, more neutral. Even when the tinnitus is more prominent, you accept it much more easily.” (NH35).

Theme 4—Untargeted Effects (n = 17)

Any other effects reported that were independent of tinnitus were classified under the theme “Untargeted effects.” This theme was derived from spontaneous comments, as the interview questions were focused on tinnitus. Participants’ comments were categorized into three subthemes describing the effects of the ST on: (1) hyperacusis, (2) hearing, and (3) other aspects.

Effects on Hyperacusis (n = 3)

It is worth reporting the improvements in sound tolerance, even though it affected only three NH participants. Notably, those same individuals had hyperacusis based on their pretrial Hyperacusis Questionnaire (HQ) scores (NH09, NH14, NH45) and did not especially exhibit quantitative improvement in their HQ scores [Table 2]. The others who also met the criteria for hyperacusis (i.e., HQ ≥ 22) did not report any effects on hyperacusis.

Effects on Hearing (n = 10)

The noise affected the hearing abilities (especially speech) of eight participants. Notably, these effects were observed regardless of the hearing status (4 NH, 4 HL). Three participants reported asking for repetitions more frequently, two reported a sensation of HL, five reported that noise masked the environment, one reported having greater difficulties to hear, and another felt that noise reduced the sound quality, creating a form of “muffled” effect. An NH participant explained their experience as “It really created an hearing loss for me when I wore them. I think it fulfilled the mission of the sound generator. I had to pay more attention to the person who was talking to me. Worse, often sounds, even ambient sounds like a little bird singing or really all the sounds around us were either non-existent or… It really caused a little deafness.” (NH04). However, two NH participants mentioned their hearing was unaffected.

Other Effects (n = 9)

Two NH participants reported experiencing small episodes of imbalances or headaches during the trial. However, on the positive side, seven participants (4 NH, 3 HL)—including the three that mentioned a hyperacusis improvement—also experienced general mental and cognitive improvements such as relaxation, concentration, and stress reduction: “White noise also helps me to concentrate. You know, for concentration, there are students who use it… I like it (…) because it greatly reduced my stress, and it helped me a lot to concentrate.” (NH09).

Theme 5—Purchasing Sound Generators (n = 29)

All participants were required to give their opinion on their choice to purchase sound generators and continue ST if they had the opportunity. The exact question was “Would you consider spending money to buy sound generators for your tinnitus? If yes, why? If not, why?.” It is important to note that participants were never informed about any cost, neither was there any intention to sell any device used in the study. All the reasons related to purchase decisions are summarized in Table 1. Regardless of the benefits provided by sound generators, among the 29 participants only three would consider purchasing generators (1 NH, 2 HL). The main reasons were that ST improved their hypersensitivity to noise, stress, and concentration. One participant also allowed themselves the opportunity to experience the “lasting” effects by trying it for a longer term. Twenty-eight others (97% of the participants) would not consider purchasing the device for various reasons, some being directly linked to and others completely independent of the tinnitus. The most frequently cited reason was the lack of effectiveness. Those participants (n = 17) explicitly mentioned that the perceived benefits were not sufficient to consider acquiring such devices. For the others, reasons specific to individuals or tinnitus were labeled “personal factors” (e.g., tinnitus severity, expectations), which seem difficult to address as they belong to individual’s physiological or psychological state. For example, several of these individuals (n = 7) found that their tinnitus was not severe or loud enough to endure the ST. This point of view is completely beyond the control of the practitioner providing the ST. Except from giving proper guidance, these aspects cannot be addressed by simply modifying some parameters of the ST. Another reason was that participants’ expectation of the therapy was to get rid of tinnitus and not to hear another supplementary noise. This aspect relates to the “cumulative” aspect of noise and tinnitus presented in Theme 3 of Table 3. The fact that participants disliked the noise was also a reason for not considering a purchase (n = 6).

The other factors were called “external factors,” as they were independent of individuals and more related to the device itself or the noise (e.g., cost, physical discomfort, aesthetic). First, two participants were hindered by the social barrier and the aesthetics of the devices. Second, even though no cost was ever discussed or mentioned by any of the interviewer or experimenters, eight participants considered the sound generators too expensive and they would not spend “such” money to acquire a device. Finally, physical discomfort was often related to glasses or to discomfort in the ear canals. Note that NH participants tended to report more personal factors (n = 19), while individuals with HL tended to report more external factors (n = 9).

DISCUSSION

A key objective of this study was to evaluate the perceived effects of ST in managing tinnitus within a relatively brief timeframe. Both improvements and worsening of tinnitus were noted and could be identified in three different domains, i.e., loudness, annoyance, and awareness of tinnitus. Interestingly, these correspond to sensory, affective, and cognitive dimensions that have briefly been documented for tinnitus, especially in the new definition of tinnitus that was recently proposed “Tinnitus is an auditory sensation (…) [and] can be lived as an unpleasant experience as a result of cognitive or emotional processes (…).”[22] These dimensions are also similar to those experienced in pain.[23]

Despite the comodulation of loudness and annoyance that participants seemed to describe in this study, it is not that clear whether the sensory and affective dimensions can be dissociated or whether they are part of a combined experience in the context of tinnitus perception. Studies have shown a relatively low correlation (0.45) between psychometric scores of tinnitus loudness and annoyance, the latter appearing to be influenced by anxiety and depression levels.[24] The dimensions of tinnitus affected by sound stimulation are still uncertain, i.e., whether the perceived change in loudness and/or annoyance are due to cognitive manipulation (such as attentional distraction from the noise) or to true sensory/physiological changes induced by the auditory stimulation. These dimensions are aligned with expectations post-ST reported in the Core Outcome Measures in Tinnitus International Delphi (COMiT’ID) study.[25] Key outcomes recommended by ≥90% of the clinical researchers involved in the study were tinnitus annoyance, awareness, and intrusiveness, along with coping strategies and ability to ignore. While some of these were implied in our interview questions (such as loudness, annoyance, and awareness, by asking “Did the level of your tinnitus vary during the study?” and “Did your tinnitus bother you during the study?”), some dimensions were spontaneously reported such as concentration and sense of control. Notably, our findings distinguished annoyance, loudness, and awareness as separate dimensions, in contrast to the Tinnitus Functional Index,[26] which includes these three aspects in the same “intrusiveness” subscale.

Our findings particularly emphasized certain adverse and unintended effects of prolonged noise stimulation that have not been reported elsewhere (i.e., tinnitus louder when removing generators, noise conflicting with previous coping strategies). However, some aspects such as concentration and sense of control showed improvement in our study. In this regard, obtaining patient input regarding the specific dimension affected by tinnitus can be valuable for tailoring treatment approaches (e.g., psychological or sound-based therapy). Moreover, integrating this information into patient education counseling, as well as discussing potential unintended effects of ST, can help realign patient expectations. Patients can indeed have unattainable expectations such as one participant who mentioned that she expected the tinnitus to completely disappear only to realize at the end of the trial that this expectation was unrealistic.

Sound Therapy Effects: A Complex Experience

Based on users’ interviews, there were several aspects in which ST played a role in changes in tinnitus, both positive and negative changes. ST appeared to have beneficial effects, primarily by (1) reducing tinnitus loudness through masking, (2) decreasing annoyance by being more tolerable, relaxing, or more pleasant than the tinnitus sound, and (3) lowering awareness by diverting attention away from it. These are in line with previously reported potential beneficial effects of ST.[27] However, despite being rarely reported in studies, a substantial number of participants also reported a worsening of their tinnitus. There are some anecdotal reports suggesting that a worsening of tinnitus may occur around 3 to 4 weeks after the start of tinnitus retraining therapy.[28] As discussed in our previous study,[16] we suspect that such aggravations in tinnitus perception might be more common than currently reported. In our study, the duration of sound generator usage was relatively short, whereas ST is usually evaluated after longer periods of use, often spanning several months.[29,30,31,32] In this context, it is hard to affirm that the observed changes would be lasting over time or if they would evolve throughout the course of therapy. An important aspect to consider is that the protocol was not designed to provide amplification for participants with HL. Improved access to sensory input could have potentially increased the benefits of the ST for the HL group. Indeed, for some patients, noise may have had a negative effect on communication, which in turn might have interfered with their perception of benefits from ST. However, there was no clear difference in the effectiveness of the sound generators between the groups as suggested by both quantitative[16] and qualitative data reported here. Nevertheless, there is a potential for harm that was reported by participants in this study that has never been reported previously. This should ignite further studies to find better predictors regarding who will benefit and who will not from ST.

Some participants had mixed feelings, reporting both positive and negative aspects of the therapy. These mixed perceptions of ST highlight the complexity of each participant’s experience and the variable nature of participation in an ST trial. One participant reported a decrease in tinnitus intensity at the beginning of the study due to the noise provided by the hearing aids, but over time the tinnitus took over from the ST (HL29). Another participant found that the sound did not completely cover his tinnitus at the beginning of the ST, which he found more bothersome than his tinnitus alone. However, when the ST was well adjusted and he no longer heard his tinnitus, it was a truly blessed moment in his life: “When you have those moments where you don’t feel like having tinnitus, it’s a pretty blessed time in your life. It makes you want to wear them more!” (NH17). For most of the participants with mixed reports about the effect of ST, these conflicting feelings were usually associated with episodes of tinnitus relief and worsening due to fluctuations in either the ST (e.g., adjustment of the sound, use of the devices) or the tinnitus itself (e.g., getting louder, tinnitus crisis). Similar fluctuations were experienced with the sound produced by the hearing aids. It makes sense that their overall perception of the ST trial was influenced by both positive and negative experiences.

Effects of the Noise: Depends on Its Characteristics?

For some participants, the noise was perceived positively—described as relaxing, pleasant, and helpful for coping with tinnitus—while others perceived it as annoying, aggressive, and unpleasant. NH participants seemed more bothered by the noise compared to HL participants, who primarily had high-frequency HL (above 3 kHz). Quantitative analyses indicated a significant difference in the perceived noise spectrum depending on participant’s hearing status. Once HL was taken into account (or considered), the resulting above-threshold level (dB SL) of the noise over the whole frequency range was higher for participants with NH (13.2 dB SL) than for those with HL (−4.6 dB SL). Additionally, a noise containing higher frequencies might be harder to integrate into environmental sounds, as suggested by Barozzi et al.,[33] while a brown (red) noise may be considered less overwhelming and more pleasant than the classical white noise. In our study, the noise spectrum was not shaped to tinnitus frequencies or to hearing threshold, which perhaps explained why participants found it annoying or perceived it as “one more sound” in addition to tinnitus.

Since tinnitus is usually in the high frequencies, it raises questions about the relevance of using a tinnitus-centered noise as recommended by Searchfield et al.[34] compared to a lower-frequency noise that could be more pleasant, or a broadband noise, which contains a wide range of frequencies. Broadband noise has been used successfully in tinnitus retraining therapy for decades to reduce tinnitus severity and has been found to yield improvement at twice as fast as counselling only.[35] Broadband noise also increases sound tolerance in adults across a wide range of frequencies.[36] However, narrow-band noise centered around the tinnitus (high) frequency interferes less with the speech frequency spectrum, allowing easier speech perception, and is more efficient in masking the tinnitus, which usually lies within the spectrum of the HL.[37] Our study also mirrored the adverse effects of broadband noise on hearing abilities, both for NH and HL participants. Audiologists could adapt the approach by choosing the “best” noise depending on their patients’ HL, preferences, and effects on speech intelligibility.

Finally, and unexpectedly, the interviews showed no clear connection between finding the noise unpleasant and worsening tinnitus, nor between finding the noise pleasant and improving tinnitus perception. This suggests that perceptions of comfort and effectiveness may rely on two distinct mechanisms, one that creates a basis/support for tinnitus coping, and another that physiologically reduces perceived tinnitus. A functional resonance imaging study showed that, after ST, local activity decreased in limbic structures and increased in auditory regions and these changes were associated with tinnitus relief.[38] Coping with tinnitus was explored in a qualitative study conducted by Munir and Pryce,[15] revealing that sound generators can help individuals by giving them a sense of validation and control over their tinnitus. This finding was also observed in our study, where sound generators provided a sense of empowerment and hope for a potential remission. In contrast, we also found that noise could conflict with initial coping strategies. Providing a noise that can support and not prevent coping is thus crucial. Finally, distinguishing one mechanism from another remains challenging and it is uncertain whether ST physiologically reduces tinnitus or simply assists with coping.

Removing Generators: Inhibition, Exacerbation, or Contrast?

A reduction in loudness or complete suppression of their tinnitus was observed when participant removed the generators. This may be similar to residual inhibition as assessed in clinical settings.[39] It is unclear from the interviews whether this perceived reduction occurred in patients whose tinnitus increased (worsened) with the generators and consequently decreased after removing them, or whether tinnitus activity was physiologically inhibited. There is a growing interest in mechanisms and predictors that may have a lasting effect of noise on tinnitus; however, conditions in which inhibition occurs require further exploration.[40,41,42] Although it is unknown whether residual inhibition as measured clinically is a potential predictor of ST efficacy, this reported short-term reduction of tinnitus seems to suggest that the noise used may desynchronize the tinnitus-dependent brain activity.[43]In contrast to the sensation of reduced loudness, some participants found that their tinnitus was exacerbated after removing generators. This could be related to a higher contrast of tinnitus in silence when compared to tinnitus in noise. Although not all participants experienced a reduction in tinnitus while wearing sound generators, it is still possible that slight reductions in tinnitus loudness went unnoticed. It is also possible that the short-term worsening was due to an exacerbation induced by the noise, and this effect could be related to an increase in the tinnitus-dependent aberrant activity, i.e., the opposite effect of residual inhibition. Exacerbation of tinnitus in humans has already been reported after exposure to a loud noise,[44] but to our knowledge, there are no reports related to a low-to-moderate levels of prolonged noise as used in sound therapies. However, a potential harmful effect of a prolonged exposure has already been discussed for humans,[45] and also shown in animals even at low intensity levels.[46] The exact conditions and mechanisms that result in worsening need to be further explored. Also, to what extent this increase in tinnitus represents a meaningful clinical worsening is unknown.

Barriers to Acquiring Sound Generators and Hearing Aids

Some of the barriers identified for the nonuse of sound generators were similar to those observed for hearing aids. Individuals often considered their tinnitus to be not severe enough to justify undertaking ST, which is consistent with a report from Linssen et al.,[10] where individuals underestimated their hearing impairment. Individuals also described hearing aids as unsightly and a social barrier, which echoes the stigma of hearing aids related to aging and reduced intelligence.[11] Auditory fatigue, discomfort related to amplification,[12] and noise stimulation have been described as exhausting or disturbing. The insufficient follow-up for hearing aids, previously reported as a barrier by Blustein and Weinstein[14] did not apply to our study due to the short duration of the trial. However, it is worth considering that extending the trial could have increased the benefits of ST. Notably, some participants would have appreciated keeping the generators for a longer period, and one was motivated to buy devices to experience the long-term effects. Finally, the high cost of hearing aids was a barrier in our study since cost was a main reason for not considering acquiring generators.

Buying Generators: Not Enough Benefits

Despite the reported benefits, the majority (97%) of participants did not consider purchasing sound generators, mainly because of a lack of perceived benefits, unchanged tinnitus severity, and cost of the devices. The lack of benefit was the most cited reason for not buying generators. Since most participants had had their tinnitus for over 10 years on average, some were already habituated to it and might have expected their tinnitus to be completely eliminated. Tinnitus patients tend to expect a complete remission or a reduction in loudness, whereas audiologists focus on promoting habituation and reducing tinnitus-related stress.[47] Counselling—not provided in this study—usually helps individuals to understand the role of therapy, and to align patients’ expectations with the possible benefits. In this context, the noise seems unhelpful since it takes away their original coping strategy and reintroduces tinnitus at the forefront of their attention.

CONCLUSION

In conclusion, our study reports qualitative information on how participants with tinnitus judged their first-time experience with sound generators after 3 weeks of use. The ST was delivered in controlled conditions, with the same duration and manufacturer for all participants to ensure consistency for all participants. Overall, our findings underline both positive and negative experiences regarding the general perception of sound generators, their effects on tinnitus, reasons that might explain the mixed results of the ST, and participants’ attitudes toward purchasing sound generators. Our study will help clinicians to present a more accurate picture of ST to their patients and help patients to better anticipate what to expect in the very first steps of their journey with sound generators.

Availability of data and materials

Data that support the findings of this study are available from the corresponding author upon reasonable request.

Author contributions

Sylvie Hébert (guarantor): Conceptualization, design, literature search, project administration, interpretation of data, manuscript editing and review, final approval of the manuscript.

Bérangère Villatte: Literature search, concept design, data analysis, interpretation of data, manuscript preparation, review and editing, final approval of the manuscript.

Charlotte Bigras: Data acquisition, data analysis, interpretation of data, manuscript preparation, review and editing, final approval of the manuscript.

Philippe Fournier: Conceptualization, data collection, interpretation of data, manuscript reviewing and editing, final approval of the manuscript.

Elizabeth Fitzpatrick: Conceptualization, design, interpretation of data, manuscript reviewing and editing, final approval of the manuscript.

Ethics approval and consent to participate

The study was approved by the Comité mixte d’éthique de la recherche du Regroupement Neuroimagerie Québec (#CMER-RNQ 06-07-001). All participants signed written informed consents before data collection.

Conflicts of interest

The authors declare that they have no conflict of interest.

Supplementary Table 1.

Additional codes not included in the qualitative analysis.

Codes	Examples
Social acceptance/Aesthetics	Discreet
	Social barrier
Physical sensation	Comfortable
	Foreign body
Use of the devices	Easy to manipulate
	Unsuited to daily routine (eg. sports, water, glasses)
Areas for improvement	Waterproof
	Adjust volume to environment
	Fit to glasses

Funding Statement

This research was funded by a doctoral studentship from the Centre Interdisciplinaire de Recherche sur le Cerveau et l’Apprentissage (CIRCA) awarded to B.V. and a grant from the Natural Sciences of Engineering Research Council of Canada (NSERC #RGPIN-2021-04397) awarded to S.H..