IntroductionBackground

Tinnitus is defined as the perception of a constant or intermittent sound without the presence of an external auditory source [,]. It is a highly prevalent symptom affecting approximately 14% of the adult population []. The daily functioning and health-related quality of life of tinnitus patients can directly or indirectly be disrupted by the distress they experience due to the tinnitus sound []. Therefore, the most commonly used and evidence-based treatments are tinnitus retraining therapy (TRT) and cognitive behavioral therapy (CBT) [,]. The high costs and time consumption of these treatments, combined with the repeated need for care of these patients, resulted in a quest to find a more cost-effective alternative [-]. As an answer, telerehabilitation is on the rise to promote self-management in tinnitus patients and reduce their dependency on in-person care. Telerehabilitation, meaning delivery of care from a distance, became a necessity rather than an opportunity during the COVID-19 pandemic [-].

The previous systematic review published in 2023 on the topic of telerehabilitation for chronic subjective tinnitus patients indicated that low-contact treatment provided from a distance through several telerehabilitation forms could be used as a substitution for or an addition to in-person clinical care for tinnitus []. It can potentially reduce the costs of treatment without sacrificing the level of effectiveness. Telerehabilitation treatment forms are easy to implement in clinical practice and form accessible treatment options []. Overall, telerehabilitation in the form of internet-based cognitive behavioral therapy (iCBT) with or without guidance, self-help manuals, self-help devices, smartphone apps, and other internet-based interventions was found to be effective in reducing tinnitus severity and distress []. Nevertheless, barriers such as lack of time, engagement, motivation, and openness of the patient need to be considered, as they can cause high dropout rates. The previous review also concluded that future research should further explore which factors are most likely to cause the lack of compliance and how clinicians can counteract these [].

Due to the advances in technology made every day, telerehabilitation will keep evolving and, therefore, remain a trending topic to follow up on. The application of digital technology in tinnitus therapy is rising stronger than ever. For example, a variety of smartphone apps for people with tinnitus are available, promising relief from the condition. Yet, it is difficult to estimate their potential [,]. After the COVID-19 pandemic, the potential benefits of telerehabilitation forms became undeniable, and recently, these new tinnitus treatment forms were even implemented in the guidelines for tinnitus care. Since then, telehealth has been a highly dynamic research field, and several new studies on telerehabilitation for tinnitus have been published in the last couple of years [-]. The high number of patients with tinnitus, their repeated need for care, long waiting lists for regular care, and the high costs related to gold-standard treatments are some of the reasons why the need to explore telerehabilitation options is so high. It might form the key to making treatments more cost-effective and more accessible without losing on a level of effectiveness. It can even benefit the effectiveness when it is used as a tool to improve tailoring of the treatment approach to the patient’s needs and as a tool to increase adherence and self-management skills of the patient, reducing the need for repeated care.

Objectives

In the past decades, various low-contact and self-management telerehabilitation treatments were developed for patients with tinnitus to search for an alternative or additional tool for the existing treatment options that are time-consuming and costly. Due to the rise in scientific research and the high number of new publications in recent years on this topic, this systematic review aimed to give an updated overview of the research concerning the effectiveness of telerehabilitation interventions for self-management of tinnitus published between 2022 and 2025.

MethodsStudy Design

This systematic review is an update of an earlier systematic review published in February 2023 []. It adheres to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines []. The predefined protocol registered in PROSPERO from the earlier review, with registration number CRD42021285450, was repeated identically.

Eligibility Criteria

The eligibility criteria can be found in . These criteria remained the same for the updated review. However, a publication date restriction was added as an exclusion criterion. Papers published before October 18, 2021, were not taken into consideration because they were already included in the previous review. The search ended on February 10, 2025.

Textbox 1. Eligibility criteria.

Inclusion criteria

Adult patients (>18 y).Subjective tinnitus as a primary complaint.A study intervention comprising any possible form of self-management or telerehabilitation.No publication date restrictions.

Exclusion criteria

No tinnitus.Objective tinnitus.Tinnitus as a secondary complaint.Full text not available in Dutch, French, German, Greek, or English.Information Sources

On February 10, 2025, a total of 5 databases were consulted: PubMed, ScienceDirect, Scopus, Web of Science, and Cochrane Library.

Search Strategy

The search queries were composed according to the PRISMA-S (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Literature Search Extension) guidelines [] (checklist provided in ). The queries comprise free terms and MeSH (Medical Subject Headings) terms for tinnitus (population) and self-management, telerehabilitation, smartphone application (intervention), and their synonyms. All terms and synonyms were controlled for value to the search query. If a term did not result in an additional number of studies found, it was eliminated to keep the search query as concise as possible while maintaining the highest level of efficiency. The search queries were not adapted from the previous review and were rerun identically, because the addition of terms did not result in more relevant eligible studies. The complete search queries can be consulted in . All searches were concluded on February 10, 2025. Besides the date restriction that papers had to be from 2022 or later, no additional search filters or restrictions were added to the search strategy. Rayyan was used to eliminate duplicates from the searches.

Selection Process

Double screening of the gathered papers was performed by 2 reviewers (Sara Demoen and EVK) independently and blindly. A first screening on the abstract and title was performed, followed by a screening on the full text. The Rayyan web platform was used to screen all papers efficiently. The final decision on the inclusion of the papers was obtained after a consensus meeting, discussing potential conflicts or papers that were labelled as maybe by at least one of the reviewers.

Data Collection

Both reviewers collected the data independently, and the data extraction tables were composed during a consensus meeting with the reviewers and the supervisors. In the earlier review, 6 subcategories were made: iCBT with guidance (feedback, monitoring, and support by a psychologist or audiologist), iCBT without guidance, self-help devices, self-help manuals, smartphone apps, and other internet-based interventions. For all of these categories, new studies were included during this updated search, except for the subcategory iCBT without guidance. For each subcategory, a data-extraction table was composed, comprising the collected study information according to the applicable items of the subdivisions: PICO (population, intervention, comparison, and outcome).

Data Items

The study and population characteristics that were listed are the author, the year of publication, the study design, the sample size (n), the gender distribution (%), the mean age and the SD (y), the type of tinnitus, the mean duration of tinnitus, and the presence of hearing loss.

Intervention is the first aspect summarized in the table of evidence. This section discusses the following items: (+ if applicable comparison: control intervention, ie, if there is a comparison with a control group [CG], these items are also discussed for the control intervention and not solely the experimental intervention): the used intervention form and specifics, such as guidance, duration, and follow-up. For the next section, Outcome, outcome measures, and dropout were presented, and finally, under the Results section, data results, GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) score, and the conclusions were listed. Each of the 5 subcategories of telerehabilitation included in this review had a table of evidence giving an overview of the information on the intervention, outcome, results, and conclusions of the included study.

Risk of Bias Assessment

The risk of bias (RoB) assessment was performed using the Cochrane RoB2 tool, a tool for assessing RoB in randomized controlled trials (RCTs) []. However, it can also be used for other study designs because the domains are, in general, also applicable []. The study design of the included studies is diverse, and for this reason, following a separate RoB assessment for each study design would not give a clear overview for comparison []. Alternative tools such as ROBINS-I (Risk of Bias in Non-Randomized Studies of Interventions) were considered, but using a different instrument in the current update would introduce heterogeneity in assessment methods. We therefore applied the Cochrane RoB tool across all designs to ensure methodological consistency with the previous version of this review, and because this tool provides a more conservative, strict assessment of bias for nonrandomized studies rather than an overly lenient one. RoBvis, an RoB visualization tool by Cochrane, was used to summarize the results in a graphic manner []. The RoB screening was performed blinded by 2 reviewers (Sara Demoen and EVK), and conflicts were resolved after a consensus meeting.

Data Synthesis

The heterogeneity in reported interventions and outcome measures precluded meta-analysis. A meta-analysis was not performed in the present review update for several methodological reasons. First, this work serves as an update of a previously published systematic review, which also did not pool results due to substantial heterogeneity. Maintaining consistency with the earlier review ensures comparability across iterations. Second, the included studies used a wide range of outcome measures that were operationalized and reported in noncomparable ways, limiting the feasibility of quantitative synthesis. Finally, the statistical reporting across studies varied considerably, with insufficiently detailed or inconsistent data formats that prevented reliable extraction of effect sizes. In combination, these factors precluded a valid and meaningful meta-analysis, and therefore, it was decided to summarize the collected data in a systematic review. This was not different in the current update.

Certainty Assessment

The GRADE framework is a tool to assess the level of certainty by examining RoB, imprecision, inconsistency, indirectness, and publication bias of the included studies []. The GRADE assessment was completed according to the GRADE handbook guidelines [].

ResultsStudy Selection

The initial database searches on February 10, 2025, resulted in 1192 potentially relevant results ( []). After removal of the duplicates and of the papers published before October 2021, a total of 496 papers remained for primary screening on title and abstract. During the primary screening, 441 papers were excluded; the reasons for exclusion are specified in . In total, 55 papers were sought for retrieval for the screening of full text. Subsequently, the screening of the full text was completed for 46 papers because the full text of 9 papers could not be retrieved. The reason for exclusion is again specified in . A total of 24 papers were found to be eligible [-]. These eligible papers were divided into the same subgroups used in the previous systematic review []. The subgroups were based on the form of telerehabilitation as an intervention: iCBT with guidance, iCBT without guidance, self-help devices, self-help manuals, smartphone apps, and other internet-based interventions.

‎Figure 1. Flow diagram for study inclusion (PRISMA 2020). Studies can belong to multiple of the listed study objective categories. iCBT: internet-based cognitive behavioral therapy; PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses. S About RoBControlled Trials

In total, 11 studies were controlled trials [,,-,,,,,,], of which 6 studies were RCTs [,,,,,]. Five RCTs had a low RoB arising from the randomization process, except for Altissimi et al [], who showed a higher RoB due to the lack of information given concerning the randomization method (D1) [,,,,,]. All controlled trials showed some reasons for concern of RoB due to deviations from the intended interventions (D2), except for Goshtasbi et al []. RoB, due to missing outcome data (D3), was scored as “high concerns” and “some concerns” for studies with respectively moderate to high dropout rates or no clearly specified plan on how to handle these missing data. All controlled studies were rated “low concerns for RoB” in the measurement of outcome (D4), except Ravi et al [], a study with a sample size of only 2 patients and only 1 questionnaire as an outcome measure. Most studies had a full, clear, correct, and specified plan for statistical analysis, meaning the concern for RoB in the selection of the reported result (D5) was low. However, Alonso-Valerdi et al [] and Ravi et al [] had a nondetailed protocol, missing information on the exact premised data analysis plan. Overall, studies were rated “some concerns” to “high concerns” of RoB. See [,,-,,,,,,] for the RoB assessment results of the controlled trials. Additionally, a summarization of RoB results can be found in .

‎Figure 2. Risk of bias in the controlled trials scoring per domain [,,,,,,,,,,]. Noncontrolled Trials

A total of 13 studies were noncontrolled trials [-,,,-,,,]. They all show some concerns about RoB due to a lack of randomization, blinding, and CGs (D1 and D2). RoB due to missing outcome data (D3) was again scored high, and there were some concerns for studies with moderate to high dropout rates. The majority of the studies specified their statistical analysis beforehand, resulting in low RoB in the selection of the reported results (D5). See [-,,,,-,,,] for the RoB assessment results of the noncontrolled trials.

‎Figure 3. Risk of Bias in the noncontrolled trials per domain [,,,,,,,,,,,,]. Certainty

In general, the included studies had a low to moderate certainty level according to the GRADE framework [].

iCBT with GuidanceStudy and Patient Characteristics

A total of 3 studies looked into the use of an iCBT program with guidance provided by an audiologist in the form of feedback, monitoring, and support through an online messaging system within the platform [,,]. All 3 studies were cohort studies with an adult study population with chronic tinnitus. Further, 2 studies additionally specified that tinnitus symptoms had to be present for more than 3 months, and it had to be at least a mild severity of tinnitus. Beukes et al [] stated that patients were only eligible if they scored at least 25 of 100 on the Tinnitus Functional Index (TFI). The mean tinnitus duration ranged from 11.64 to 13.91 years between the studies. Beukes et al [] was the only study controlling for the presence of hearing loss, while Beukes et al [] was the only one specifying the tinnitus characteristics, namely the tinnitus location. The total sample is 391 patients receiving iCBT with guidance. For further specifications, see Table S1 in . Two studies examined an iCBT program with minimal guidance [,]. However, it needs to be noted that both these published studies were from the same author and concerned the same sample data. Both studies used data from a cohort of 228 patients with long-standing chronic tinnitus with a duration of 17.68 (SD 19.42). For further specifications, see Table S1 in .

Results of Individual Studies

All 3 studies used a similar iCBT program consisting of an interactive e-learning program with 22 modules based on the original CBT self-help program developed by Andersson and Kaldo [] (2004) [,,]. The interventions were guided by an audiologist who monitored progress, provided feedback on worksheets that were completed, outlined the content of new modules, and answered questions. The guidance by the audiologist was offered through a 2-way encrypted online messaging system. The primary tinnitus outcome of all studies was tinnitus severity or distress measured by the TFI. Various secondary outcome measures, using self-report questionnaires, were appended. All studies used data from previous trials. See for further details. All studies concluded that the iCBT with guidance provided by an audiologist trained in iCBT principles was effective in the treatment of chronic tinnitus. Beukes et al [] demonstrated that iCBT has long-term positive effects, and Manchaiah et al [] added that it can help individuals with tinnitus to think more positively by changing unhelpful patterns. Both studies from Rodrigo et al [,] used the same data that was gathered from multiple trials and looked into the use of an iCBT program with minimal guidance through an encrypted messaging system solely for support. This iCBT program was again based on the original CBT self-help program developed by Andersson and Kaldo [] (2004). However, it was adapted to a set of only 21 modules, of which 5 were optional. These 2 studies also found a decrease in TFI after completing the iCBT program and additionally investigated predictors of iCBT outcomes. For more information, see .

Table 1. Table of evidence for the studies concerning iCBTa with guidance.StudyInterventionOutcomeResultsAuthor and yearInterventionTreatment periodFollow-upPrimary tinnitus outcome measureAdditional outcome measuresStatistical analysis methodData resultsStatistical analysis resultsDropoutConclusionGRADEbBeukes et al [], 2022Interactive e-learning iCBT consisting of 22 modules and guidance by an audiologist8 weeksFU1c: postintervention assessment; FU2d: 2-months postintervention; FU3e:1-year postinterventionTFIfTHSg.Effect sizes, LMMh, and the RCIi.Significant BLj-FU3 TFI mean improvement of 24.3 (SD 22.5; T0: 54.23, SD 20.7; T3: 30.15, SD 24.51). There was a clinically significant change for 56 (42%) of the participants using the reliable change criterion of 26.98 at T3.Statistically significant BL-FU3 difference with a large effect size (Cohen d=1.06, CI 0.80 to 1.32).68/200 (34%) did not complete FU1, 83/200 (43%) FU2, and 68/200 (34%) FU3.The study provides evidence of long-term efficacy for an audiologist-guided iCBT tinnitus intervention.⊕⊕⊕○Beukes et al [], 2022Interactive e-learning iCBT consisting of 22 modules and guidance by an audiologist8 weeksFUk=2 monthsTFIGAD-7l, PHQ-9m, ISIn, EQ-5D-5L, THS, TCQo, tinnitus and hearing survey.Pretest and posttest effect size for all primary and secondary outcomes; 2-sided P values using α=.05.A significant, large effect size TFI change postintervention (mean 1.60, 0.91-2.23), which was maintained at FU.The RCI indicated a prepost score difference of 19.51 on the TFI would be a clinically significant change.4/27 did not complete FU1 and 9/27 did not complete FU2.This pilot study has indicated the feasibility of iCBT for tinnitus in the United States.⊕⊕⊕○Manchaiah et al [], 2024iCBT with guidance by an audiologist8 weeksFU: posttreatmentTFIGAD-7, PHQ-9, ISI, EQ-5D-5L.Mixed method approach, paired sample t test as the data met the assumption of normality using the Shapiro-Wilk test (P>.05).More participants reported positive experiences following the intervention in the overall sample (preintervention mean 0.78, postintervention mean 1.63, P<.001).53/164 (32.3%) provided at least 1 positive experience in both pre and postintervention (group 1), 19/164 (9.1%) provided positive experiences only during preintervention, 49/164 (29.9%) provided positive experiences only during postintervention, and 28.7 participants did not provide any positive experience in both pre- and postintervention.NRp.iCBT can help individuals with tinnitus to think more positively by changing unhelpful thought patterns.⊕⊕⊕○Rodrigo et al [], 2022Interactive e-learning iCBT consisting of 21 modules and minimal guidance by an audiologist8 weeks (2-3 modules a week)FU1=postintervention assessment; FU2=2-months postinterventionTFIISI, PHQq, HHIA-Sr, HQs, CFQt, SWLSu.Paired t test.The mean BL tinnitus severity and tinnitus severity following the iCBT intervention were 57.93 (SD 19.17) and 34.22 (SD 22.76), respectively, indicating a significant decrease (P<.001).65% had a successful iCBT outcome, meaning they had a 13-point or higher reduction in TFI after the intervention.NR.Some predictors can be noted for a better outcome of iCBT, such as, for example, level of education, BL tinnitus, workload, and disability allowance.⊕⊕⊕○Rodrigo et al [], 2021iCBT8 weeks (2-3 modules a week)FU: postinterventionTFIISI, PHQ, HHIA-S, HQ, CFQ, SWLS.Paired sample t tests.Undertaking iCBT significantly reduced tinnitus severity scores (t227=16.37; P<.001) from a mean BL severity score of 57.93 (SD 19.17) to a mean post-iCBT severity score of 34.22 (SD 22.78), as measured by the TFI.A clinically significant 13-point change in TFI scores was achieved by 150 of the 228 participants after the intervention.NR.Decision tree models, especially the CARTv and gradient boosting models, appeared to be promising in predicting iCBT outcomes.⊕⊕⊕○

aiCBT: internet-based cognitive behavioral therapy.

bGRADE: Grading of Recommendations, Assessment, Development, and Evaluation.

cFU1: first follow-up moment.

dFU2: second follow-up moment.

eFU3: third follow-up moment.

fTFI: tinnitus functional index.

gTHS: tinnitus handicap scale.

hLMM: linear mixed effects models.

iRCI: Reliable Change Index.

jBL: baseline.

kFU: follow-up moment.

lGAD-7: Generalized Anxiety Disorder-7.

mPHQ-9: Patient Health Questionnaire-9.

nISI: Insomnia Severity Index.

oTCQ: Tinnitus Cognitions Questionnaire.

pNR: not reported.

qPHQ: Physical Health Questionnaire.

rHHIA-S: Hearing Handicap Inventory for Adults-Short Form.

sHQ: Hyperacusis Questionnaire.

tCFQ: Cognitive Failures Questionnaire.

uSWLS: Satisfaction With Life Scale.

vCART: Communication Access Realtime Translation.

Self-Help DevicesStudy and Patient Characteristics

Three controlled trials, of which 2 RCTs, studied the use of self-help devices for the treatment of chronic tinnitus [,,]. The studies had a total sample size of 286 patients. The specifications of all 3 studies can be consulted in Table S2 in .

Results of Individual Studies

All 3 studies looked into the use of a self-help device delivering a form of sound therapy. For Altissimi et al [], this was the use of a sound generator for TRT treatment, and this was the CG. The intervention group received TRT treatment using a smartphone app. Searchfield and Sanders [] also used a smartphone app in combination with the self-help devices for sound therapy, and the CG did as well use an app, but without a self-help device. All studies concluded that the use of a self-help device was effective in reducing tinnitus distress according to the TFI or Tinnitus Handicap Inventory [,,]. Further details can be found in .

Table 2. Table of evidence for the studies concerning self-help devices.StudyInterventionOutcomeResults Author and yearInterventionControl groupTreatment periodFollow-upPrimary tinnitus outcome measureAdditional outcome measuresStatistical analysis methodData resultsStatistical analysis resultsDropoutConclusionGRADEaAlonso-Valerdi et al [], 2021(1) TRTb; (2) TEAEc; (3) ADTd; (4) BBTe; (5) musicNo treatment.2 months (60 min daily)FU1f: after 1 week; FU2g: after 5 weeks; FU3h: after 8 weeksTHIiHADSj, audiometry, tinnitus characteristics analysis, and BLk neurophysiological analysisIdentification of ± or 0 effects t testAll negative differences between BL and FUl were statistically significant (P=.009)29.4% of the TRT group reported diminution of their tinnitus perception, followed by those in TEAE (18.8%), BBT or ADT (14.3%), and music (7.7%) groups.Dropout: music 6/16, BBTm 4/18, TRT 3/18, TEAE 5/18, ADT 5/18, and control 3/14TRT most effective sound-based therapy to reduce tinnitus perception for individuals without anxiety. BBT, ADT, and TEAE produced very similar effects.⊕⊕⊕○Altissimi et al [], 2024TRT with mobile appTRT with a traditional sound generator.6 monthsFU1: after 3 months of treatment; FU2: after 6 months of treatmentTHITRT interview: with questions on tinnitus, hearing loss, hyperacusis, classifying gravity, the annoyance, and the effect on the QoLn of tinnitus, using a VASo.Descriptive data by counts of a mixed-effect model Tukey testA significant difference in the THI scores was found in category 1 patients between those treated with a sound generator vs app (–14.529, P<.001)Improvement was found in category 0 patients, both treated with the app and sound generator; conversely, in category 1, significant improvement was seen only for the group using a sound generator.No dropoutTRT can be delivered to patients with mild tinnitus (class 0) through mobile apps with results comparable to traditional sound generators.⊕⊕⊕○Searchfield and Sanders [], 2022The digital therapeutic consisted of an app for iPhone or Android smartphone, Bluetooth bone conduction headphones, a neck pillow speaker, and a cloud-based clinician dashboard to enable messaging and app personalizationThe control app was a popular self-help passive sound therapy app called WNp Lite.12 weeksFU1: midtrial=6 weeks; FU2: postintereventionTFIqCOSITr, SUSs, MAUQtANOVA with 2 groups and 4 repeated measures, intent-to-treat and completer (per protocol) analysesMean changes TFI for USLu at 6 (16.36, SD 17.96 points) and 12 weeks (17.83, SD 19.87 points) were clinically meaningful (>13 reduction), the mean changes in WN group were not (6 wk 10.77, SD 18.53 points; 12 wk 10.12, SD 21.36 points)A statistically higher proportion of USL participants achieved meaningful TFI change at 6 weeks (55%) and 12 weeks (65%) than the WN group at 6 weeks (33%) and 12 weeks (43%).4/35 USL and 6/36 WNThe USL group demonstrated a higher proportion of responders than the WN group. The usability of the USL therapeutic was similar to the established WN app. The digital polytherapeutic demonstrated significant benefit for tinnitus reduction, supporting further development.⊕⊕⊕○

aGRADE: Grading of Recommendations, Assessment, Development, and Evaluation.

bTRT: tinnitus retraining therapy.

cTEAE: therapy for an enriched acoustic environment.

dADT: auditory discrimination therapy.

eBBT: broad band therapy.

fFU1: first follow-up moment.

gFU2: second follow-up moment.

hFU3: third follow-up moment.

iTHI: Tinnitus Handicap Inventory.

jHADS: Hospital Anxiety and Depression Scale.

kBL: baseline.

lFU: follow-up moment.

mBBT: Broad band therapy.

nQoL: quality of life.

oVAS: visual analog scale.

pWN: white noise.

qTFI: Tinnitus Functional Index.

rCOSIT: Client-Oriented Scale of Improvement in Tinnitus.

sSUS: System Usability Scale.

tMAUQ: mHealth App Usability Questionnaire.

uUSL: UpSilent app.

Self-Help ManualStudy and Patient Characteristics

Only 1 cohort with a sample size of 10 adult patients with tinnitus and with bilateral mild sloping sensorineural hearing loss who wore a hearing aid used a self-help manual as an intervention []. Further details can be found in Table S3 in .

Results of Individual Studies

Shetty [] examined the use of a self-management manual for tinnitus with content such as (1) hearing aids; (2) listening to songs or soothing sounds; (3) tips for food intake; (4) exercise, yoga, and meditation; (5) locking negative thinking; (6) distraction techniques; (7) progressive muscle relaxation; (8) slow breathing; and (9) attention control and imaginary techniques. The self-help manual was prepared using the health literacy recommendations by adapting the United States Department of Health and Human Services. It was concluded that the self-management manual effectively reduces tinnitus perception and associated problems []. See for further information.

Table 3. Table of evidence for the studies concerning the self-help manual.StudyInterventionOutcomeResultsAuthor and yearInterventionControl groupTreatment periodFollow-upPrimary tinnitus outcome measureAdditional outcome measuresStatistical analysis methodData resultsStatistical analysis resultsDropoutConclusionGRADEaShetty et al [], 2024Self-management manual for tinnitusN/Ab30 daysFUc: postinterventionTHId, TFIeN/AIndependent samples t test.The results revealed a significant reduction in THI after 30 days of manual use (t9=1.99, P=.07). A similar result was found in TFI, in which functional impairment from tinnitus was significantly reduced between visits one and two (t9=6.13, P=.001)The means of THI and TFI scores in visit 1 were 31 (SD 5.16) and 58 (SD 5.92), respectively. Furthermore, at visit 2, the THI and TFI scores were 29 (SD 4.59) and 45 (SD 7.05), respectively.No dropoutPatients who have used the manual reported relaxing and coping with tinnitus. It is concluded that the self-management manual effectively reduces tinnitus perception and associated problems.⊕⊕⊕○

aGRADE: Grading of Recommendations, Assessment, Development, and Evaluation.

bN/A: not applicable.

cFU: follow-up moment.

dTHI: Tinnitus Handicap Inventory.

eTFI: Tinnitus Functional Index.

Smartphone AppsStudy and Patient Characteristics

A total of 13 articles, with a combined sample size of 23,637 participants, discussed the use of a smartphone app as part of the treatment of tinnitus patients [,,,,-,-,-]. Clarification on the study and patient characteristics of the studies examining the use of a smartphone app is noted in Table S4 in .

Results of Individual Studies

The intervention used a smartphone app in 13 studies [,,,,-,-,-]. First, in 8 studies, this smartphone app was used to deliver a form of sound therapy to the patients. These studies showed that sound therapy through a smartphone app can be beneficial in reducing tinnitus distress. In 2 of these studies, Goshtasbi et al [] and Searchfield and Sanders [], the smartphone app combined both sound therapy and CBT or psychoeducation, respectively. Second, 5 studies look into TRT, CBT, or psychoeducational treatment delivered through a smartphone app. TRT with a mobile app appeared to be as effective as traditional TRT with or without a sound generator, and daily training using a CBT-based app targeting maladaptive beliefs may decrease tinnitus intrusiveness and handicap. Psycho-education through a smartphone app gave promising results in reducing tinnitus distress, but did not result in changes in tinnitus loudness. Finally, 1 study investigated the use of a smartphone app in the treatment of a specific subpopulation of tinnitus patients, namely patients with somatic tinnitus []. This pilot study provides evidence that an app-based physiotherapy intervention with a cervical spine exercise program could be an effective option for treating somatic tinnitus []. See for an overview of the detailed information for all studies concerning treatments using smartphone apps.

Table 4. Table of evidence for the studies concerning smartphone apps.StudyInterventionOutcomeResultsAuthor and yearInterventionControl groupTreatment periodFollow-upPrimary tinnitus outcome measureAdditional outcome measuresStatistical analysis methodData resultsStatistical analysis resultsDropoutConclusionGRADEaAltissimi et al [], 2024TRTb with mobile appTRT with a traditional sound generator6 monthsFU1c: after 3 months of treatment; FU2d: after 6 months of treatmentTHIeTRT interview: with questions on tinnitus, hearing loss, hyperacusis, classifying gravity, the annoyance, and the effect on the QoLf of tinnitus, using a VASg.Descriptive data by counts, a mixed-effect model Tukey test.A significant difference in the THI scores was found in category 1 patients between those treated with a sound generator vs app (–14.529, P<.001), while no significance was found in category 0 patients (–1.186, P=.78).Improvement was found in category 0 patients, both treated with the app and sound generator; conversely, in category 1, significant improvement was seen only for the group using a sound generator.No dropout.TRT can be delivered to patients with mild tinnitus (class 0) through mobile apps with results comparable to traditional sound generators.⊕⊕⊕○Goshtasbi et al [], 2025Smartphone-based CBTh and customized sound therapy.Waitlisted8 weeks (1 weekly CBT module and 2 h daily sound therapy)PosttreatmentTFIiPHQ-9j, GAD-7k, PSQIl, PSSm.Independent sample t test and chi-square tests of independence.Subjects in the IGn showed greater improvements on their TFI (16.7, SD 14.9 versus 1.9, SD 10.8 (P<.001) and on 7 of 8 subscales (P <.001-.002).18 (38.3%) IG patients showed a TFI improvement of 20 or more points posttreatment.15 patients dropped out between week 2 and week 8 and were not included in the IG.Smartphone-based CBT, providing CBT and customized sound therapy, was effective in reducing symptom severity and improving anxiety, sleep, and mood.⊕⊕⊕○Kutyba et al [], 2022Free ReSound Tinnitus Relief appN/Ao6 months (≥30 min/day)FU1: after 3 months of treatment FU2: after 6 months of treatmentTHI, TFIUser survey.Repeated measures ANOVA with Bonferroni correction.THI: F2,102=62.4; P<.001; e2=0.550, TFI: F2,102=28.9; P<.001; e2=0.362.In the study, the general severity decreased after the first 3 months (THI: –6.5 in mean score in comparison to BLp; TFI: –6.3 in mean score) and again in the following 3 months of using the application (THI: –19.4; TFI: –14.3).NR.The sound therapy through the use of a smartphone app appeared to be beneficial in reducing subjective tinnitus distress.⊕⊕⊕○Kutyba et al [], 2022Free ReSound Tinnitus Relief appParticipants not willing to participate in the intervention group6 months (≥30 min/day)FU1: after 3 months of treatment FU2: after 6 months of treatmentTHIVAS, user survey.Independent (unpaired) t test or a chi-square test, and a mixed-design ANOVA with Bonferroni adjustment for multiple comparisons.After 3 months of using the app, the THI global score significantly decreased (P<.001) in the study group, decreasing again at 6 months (P<.001).A clinically important change (>20) in the THI was reported by 39% (17/44) of the study group and in none of the controls.Dropout rate: 53% (50/94) in the study group and 55% (29/53) in the control group.Use of an app that generates background sounds appears to be an effective way of reducing tinnitus severity.⊕⊕⊕○Michiels et al [], 2024App-based cervical spine exercise programWaiting list9 weeksFU: posttreatmentTHI miniTQqN/A.Independent sample t test and chi-square tests of independence, and mixed model ANOVA.Reduction in THI and miniTQ for the IG, not for CGr. Strong significant decrease in THI (from mean 51.33, SD 23.77 to mean 38.96, SD 21.66) and miniTQ (from mean 13.27, SD 5.81 to mean 9.14, SD 5.27).Effect sizes for THI were d=1.71 (95% CI 0.85 to 2.56) and for miniTQ d=1.02 (95% CI 0.24 to 1.80).CG: 3 dropouts; IG: 4 dropouts.This pilot study provides evidence that an app-based physiotherapy intervention could be an effective option for treating somatic tinnitus.⊕⊕⊕○Oron et al [], 2022Brief cognitive behavioral training for tinnitus relief using a mobile GGTIs appN/A16 days (3 min/day)FU1: at level 24 half way through the app; FU2: level 48 final level of the appH-THItVAS, user survey.Repeated measures ANOVA of completers.A significant linear trend was present, F1,13=14.5, P=.002, d=2.12. Oron et al [] indicated 5 was a steady decrease in the H-THI score of 28.57, 22.14, and 22.0 for BL, FU1, and FU2, respectively.50% of completers have shown reliable change (indicated by their RCIu).12/26 (46%).Several minutes a day of training using a CBT-based app targeting maladaptive beliefs may decrease tinnitus intrusiveness and handicap.⊕⊕⊕○Ravi et al [], 2025Mobile-based TRTConventional face-to-face TRT3 months (2×/week)FU1: after 1 month of treatment; FU2: after 2 months of treatment; FU3: after 3 months of treatmentTHIN/A.N/A.Patient A: BL=40, FU1=40; FU2=36; FU3=24. Patient B: BL=36, FU1=24; FU2=16; FU3=14.N/A.N/A.Mobile-based TRT might be a good option for patients who cannot attend in-person sessions.⊕⊕○○Schlee et al [], 2022Educational smartphone appN/A4 monthsFUv: postinterventionTHI, TSw NRSxN/A.Wilcoxon Signed Rank Test.Statistical significance between the pre- and postintervention phase (W=345, P=.02).Improvements on the tinnitus severity NRS reached an effect size of 0.408, while the improvements on the THIy were much smaller with an effect size of 0.168.26 of 52 (50%) participants dropped out.Educational training on a smartphone app without the guidance of a medical person can result in small to medium improvements in tinnitus distress measures, but not for tinnitus loudness.⊕⊕⊕○Searchfield et al [], 2022The digital therapeutic consisted of an app for iPhone or Android smartphones, Bluetooth bone conduction headphones, a neck pillow speaker, and a cloud-based clinician dashboard to enable messaging and app personalizationThe control app was a popular self-help passive sound therapy app called WNz Lite.12 weeksFU1: midtrial=6 weeks FU2: postintereventionTFICOSITaa, SUSab, MAUQacANOVA with 2 groups and 4 repeated measures intent-to-treat and completer (per protocol) analysesMean changes in TFI for the USLad group at 6 (16.36, SD 17.96 points) and 12 weeks (17.83, SD 19.87 points) were clinically meaningful (>13 points reduction), the mean change in WN scores were not clinically meaningful (6 weeks 10.77, SD 18.53 points; 12 weeks 10.12, SD 21.36 points).A statistically higher proportion of USL participants achieved meaningful TFI change at 6 (55%) weeks and 12 (65%) weeks than the WN group at 6 (33%) weeks and 12 (43%) weeks.4/35 USL and 6/36 WN.The USL group demonstrated a higher proportion of responders than the WN group. The usability of the USL therapeutic was similar to the established WN app. The digital polytherapeutic demonstrated significant benefit for tinnitus reduction supporting further development.⊕⊕⊕○Seol et al [], 2023Sound therapy smartphone appN/A6 monthsFU1: midtrial=3 months; FU2: postintereventionTHIBDIae, SSLaf, VAS stress assessment, hearing tests.1-way RMag ANOVA as it passed the normality test (Shapiro-Wilk).The THI scores, on the other hand, significantly decreased over the 6-month period (P=.02); when the Bonferroni correction method was applied for all P values, statistical significance was observed for THI scores between visits 1 and 3 (P=.006).After 6 months of listening to the sound stimuli, participants’ perceived tinnitus handicap severity significantly decreased.NRah.The findings of this study demonstrate the potential benefit of the tinnitus application for tinnitus improvement.⊕⊕⊕○Suh et al [], 2023Smart device–based TRT with smart pad appConventional TRT3 months (3 sessions 30-60 min) with 1 month intervalFU1=postintervention assessment; FU2=2-months postinterventionTHIVAS, STAIai, BDI, PSQI, pure tone audiometry.t tests and chi-square tests.The greatest THI improvement at 3 months was –23.3 (95% CI –33.41 to –13.4), and at 2 months –16.8 (95% CI –30.8 to –2.8), and the reductions were significant, P<.001.There was no difference between IG and CG (P<.001).NSaj.TRT based on smart devices can be an effective alternative to conventional in-person TRT.⊕⊕⊕○Tang et al [], 2022FTRSak mobile appCohort2 h/dFU every 3 monthsTHIHADSal, AISam, VAS, FTQan, TCSao.Descriptive statistics.N/A.N/A.Only 22.9% of patients adhered to the application (29,864 patients were excluded because of incomplete records of personal and tinnitus-related information).FTRS greatly reduced medical costs in comparison to traditional in-person care and enabled patients with tinnitus to arrange their own treatment times.⊕⊕⊕○Tang et al [], 2025DCRSap through a smartphone appUnmodified music3 months (2 h/d)FU1: 1 month of treatment; FU2: 2 months of treatment; FU3: 3 months of treatmentTHIHADS, AIS, FTQ, TCS.Linear mixed models and a multiclass logistic model with a stepwise function.DCRS-treated patients after 3 months of treatment had significant tinnitus relief compared to those in the CG group (P<.001).At 3 months, 92.5% of patients undergoing DCRS reported tinnitus relief or disappearance.NS.DCRS showed to be a promising noninvasive therapy for chronic tinnitus that can be delivered through a mobile app.⊕⊕⊕○

aGRADE: Grading of Recommendations, Assessment, Development, and Evaluation.

bTRT: tinnitus retraining therapy.

cFU1: first follow-up moment.

dFU2: second follow-up moment.

eTHI: Tinnitus Handicap Inventory.

fQoL: quality of life.

gVAS: visual analog scale.

hCBT: cognitive behavioral therapy.

iTFI: Tinnitus Functional Index.

jPHQ-9: Patient Health Questionnaire-9.

kGAD-7: Generalized Anxiety Disorder-7.

lPSQI: Pittsburgh Sleep Quality Index.

mPSS: Perceived Stress Scale.

nIG: intervention group.

oN/A: not applicable.

pBL: baseline.

qminiTQ: Mini Tinnitus Questionnaire.

rCG: control group.

sGGTI: GG Tinnitus.