Inclusion criteria.

Participants must meet all the following criteria:

1. Aged 18 years or over
2. Experienced tinnitus symptoms for ≥ = 3 months, with tinnitus self-assessed as impacting quality of life
3. Have access to a smartphone
4. Able to speak and read English to a sufficient level
5. Able and willing to give consent for the study prior to participation

Exclusion criteria.

Participants meeting any of the following criteria will not be eligible for inclusion:

1. Tinnitus with potential “red flag” symptoms (e.g., unilateral or pulsatile tinnitus)
2. Significant mental health problems e.g. history of suicidal ideation or requirement for psychiatric/psychological support beyond Primary Care level
3. Have required hospitalisation for mental illness or currently taking an antipsychotic drug
4. Currently taking part in another clinical trial for hearing/tinnitus
5. Awaiting surgical intervention for hearing/tinnitus
6. Previously undergone tinnitus therapy of any type e.g. CBT or TRT

Additional Note: Patients with hearing aids are eligible to participate in the study

Pregnant or breastfeeding women (for insurance reasons).

Oto smartphone tinnitus treatment programme.

The Oto app is a digital therapeutic delivered via smartphone. Participants receive personalised tinnitus therapy based on their interaction with the app, with therapy combining CBT with mindfulness, patient education, cognitive exercise and physical therapy (physical stretches/exercises). Participants will be offered a series of daily therapy sessions, where the user listens to recorded audio content that takes them through elements of tinnitus therapy. Participants will work their way through progressive modules using a spiral curriculum [15], where the sessions build on techniques and exercises learned in previous sessions. The programme lasts approximately 6 weeks, using the app 4–5 times per week. Users can personalise their therapy by listening to additional modules and sounds from the sound library. The intervention will begin immediately following randomisation.

One-to-one conventional tinnitus therapy.

The control arm is designed to represent standard care intervention and was developed based on relevant guidance (BSA Practice Guidance Tinnitus in adults 2021 and NICE guidelines (NG-155)) and via consultation with clinicians providing NHS tinnitus therapy. Participants will undertake one-to-one tinnitus therapy from a trained audiologist / hearing therapist / psychologist via video call.

A detailed specification for the control arm intervention can be found in S2 File. In summary, an initial 60 minute session will incorporate a detailed audiological history, an explanation of tinnitus, and therapy tailored to the participant’s needs and understanding, discussing options for treatment and rehabilitation. Questionnaires will be administered as appropriate to assess tinnitus handicap, anxiety and depression in order to establish the functional impact of tinnitus and guide appropriate management. If appropriate, onward referral will be made to other specialties. An individual management plan will then be formulated and agreed with the participant and may include: Relaxation therapy, CBT, Signposting for sound enrichment and relaxation, Recommendation for hearing assessment and/or aid fitting tinnitus maskers where relevant. Signposting will be provided as relevant to trusted resources and local wellbeing services, or to medical care for depression or anxiety.

The control intervention will consist of a minimum of the initial 60 minute tinnitus therapy session, with further 30–60 minute follow up sessions at the discretion of the participant and therapist, up to a maximum of six sessions.

A panel of five therapists will deliver the therapy, selected via interview, and will have training and experience in the treatment of patients with tinnitus using the techniques specified for the control intervention. Participant questionnaires will assess participant perspective of the therapy and will be used for quality assurance.

One-to-one therapy appointments will be offered immediately following randomisation, with the expectation that the first session will occur within 7 days of randomisation. In order to facilitate retention, participants in the control group will be offered access to the Oto App following final data collection at the end of the 12-month trial period.

Intervention adherence.

Engagement with the Oto app by participants will be recorded, including frequency of use, overall duration of use and number of modules completed. Engagement with one-to-one therapy will be assessed via therapist reporting, in terms of therapy session dates and number completed for each participant. The number of participants actively using the interventions at 3 and 6 months will be assessed.

At enrolment, participants will be directed to contact their GP for hearing aid referral to a rapid access service such as is provided by accredited ‘Any Qualified Provider’ (AQP) high street audiologists. Any form of hearing aid provision including tinnitus maskers will be permitted and this will be recorded via participant questionnaire at assessments.

No medication is administered within the study, nor are any medications prohibited. Concomitant tinnitus therapy, therapist-led or digital, is not permitted and protocol violations will be screened for at all follow up assessments.

Primary outcome.

The primary outcome is the difference of the Tinnitus Functional Index (TFI) [16] total score measured at 6 months following allocation and at baseline. The TFI is a 25-item participant-completed questionnaire that is sensitive to changes in tinnitus severity and impact [16], and it is the tool recommended by NICE for this purpose [6]. Participants will be administered the questionnaire via the trial’s electronic data capture system using a smartphone or tablet at baseline, and then at 4,12, 26 and 52 weeks.

Secondary outcomes.

Adverse events will be recorded in both arms from randomisation until and including 12 weeks, captured via participant self-report and structured items in the electronic participant questionnaires.

Two health-related quality of life measures will be used to inform the health economic evaluation: The EuroQol 5D (5 level)(EQ-5D-5L) [17, 18] and Health Utilities Index Mark 3 (HUI3) [19] The EQ-5D-5L is currently the preferred preference-based instrument to measure HRQoL in adults by NICE, and the HUI3 is particularly well-suited to hearing-related research [20]. Both will be used as the basis for a health economic assessment (see ‘analyses’).

Other assessments.

The System Usability Scale (SUS) [21] has become an industry standard for app assessment, and participants in the Oto app intervention arm will be asked to complete this at 12 week follow up. Similarly, the control group will be asked to complete a questionnaire to assess their experience of one-to-one therapy.

Pre-screening.

Potential participants will be directed to the trial web page, where they will be informed about the trial aims and design. If interested in taking part, they will complete an online pre-screening form to assess their eligibility for the trial. If potentially eligible, the Patient Information Sheet (PIS) and Informed Consent Form (ICF) will be emailed directly to the participant, as well as a video explaining the trial. If interest is sustained, the potential participant will be asked to book a telephone/video call with a member of the central trial team.

Screening and eligibility confirmation.

Screening will be conducted remotely by a member of the central trial team, during the same telephone/video call where informed consent is obtained. A member of the central trial team will use medical history obtained from the participant to confirm that the individual meets the inclusion and exclusion criteria. A record of screen failures that do not meet the inclusion/exclusion criteria will be retained.

Informed consent.

Prior to consent, the potential participant will have had the adequate time and opportunity to read all information materials and speak to a suitably qualified and delegated member of the central trial team. If the participant is willing to provide informed consent this will be documented via signature on an electronic consent form (S4 File), and the file will be securely stored. The participant’s general practitioner will be informed of their involvement in the trial.

Identifiable participant details (confidential details, including name and date of birth) will be held in a database separated from the research database after receiving participant consent. The research database will never hold personal identifiable information. Automatic reminders will be sent to participants by doing a one-time recall of the identifiable information by matching the unique study ID.

Hearing screening.

Participants will be asked to have headphones available at the time of their screening and consent phone call. Participants who consent to trial involvement will be directed to download a smartphone application to screen their hearing for the purpose of stratification at randomisation. Participants with an iOS-based device will be asked to download the ‘Easy Hearing Test’ application (produced by Hiroaki Ito). Those with an Android-based phone will be asked to download the ‘Hearing test’ app (produced by e-audiologica). The participant will then share the hearing test audiogram result with the trial team who will record the mean pure tone threshold (averaged at 0.5, 1, 2 and 4kHz). All participants will be instructed to seek formal audiometric testing via their general practitioner. These screening applications have been validated in the home environment (publication under peer review).

Randomisation.

A member of the central trial team will inform the participant of their randomisation allocation during their eligibility assessment, and this will be followed up by an email confirmation. Randomisation represents ‘day 0’ of the trial and outcome assessments will be timed from this point.

Baseline measurements- Day 0 (+7 days).

Prior to starting the intervention, participants will report baseline measurements in their survey which will include detailed documentation of tinnitus and audiological history and treatments, as well as completion of trial outcome measures: TFI questionnaire and EQ-5D-5L and HUI3 questionnaires.

Administration of interventions.

Following baseline data collection, the participant will be granted access to their allocated intervention, with immediate access to the Oto app, or immediate access to booking their first therapist appointment within the following seven days.

Follow-up week 4 (±7 days).

The first follow up for data capture will occur at week 4, where participants will be required to complete an online TFI questionnaire. They will be contacted by a member of the trial team to respond to questions about adverse events and any changes in medication or the introduction or modification of hearing devices (e.g. aids).

Follow-up week 12 (±14 days).

Further online data capture will occur at week 12, with online TFI questionnaire, SUS and EQ-5D-5L and HUI3 questionnaires. Additional online questionnaires will record any primary and community healthcare appointments and A&E visits related to tinnitus and any personal costs incurred, appointments attended or missed work. A further investigator phone call will cover adverse events and any medication or devices changes.

Follow-up week 26 (±14 days) and week 52 (±28 days).

Data capture at 26 weeks will include all data collected at week 12, minus adverse event and SUS data.

Data capture.

For the duration of the trial, participants will be asked to record their responses to questionnaires directly into the Lindus Health Electronic Data Capture (EDC) platform, via a participant survey link, which will be emailed and/or texted to them.

Statistical methods.

The primary analysis will be conducted on the observed values of the per-protocol population. In addition to this, as a sensitivity analysis a complete-cases, a per-protocol, and an intention-to-treat (ITT) analysis with imputed values will be completed. Compliance will be defined as a minimum of 60 minutes of total use for the Oto app arm, and one or more sessions for the one-to-one therapy group. Participants with poor compliance will be recorded and these participants will be included in the ITT analysis but not the per-protocol analysis.

The primary outcome will be reduction in total TFI score at the 6 month point in the per protocol population. Non-inferiority will be assessed by constructing a 95% confidence interval around the mean difference in TFI reduction between the intervention arms, comparing the upper bound of the confidence interval to the pre-specified non-inferiority margin.

A secondary analysis of the primary outcome will examine significant differences between the control and intervention group’s TFI score at 6 months using a general linear model or a comparable suitable method, using the stratification variables and other potential confounding variables as covariates.

Secondary outcomes will be compared between groups and over time with mixed-effects regression models for repeated measures, using the stratification variables and other potential confounding variables as covariates. Logistic models will be used for binary outcomes, and Gaussian models for continuous outcomes. An interim analysis may be conducted once a sufficient number of participants have completed the 6 month milestone. The results of the interim analysis will not influence the continuation of the trial, and will not be disclosed beyond the trial management group.

Handling of missing and incomplete data.

The EDC platform provides real-time notifications to the trial team to inform them of any missing data to minimise the risk of missing data. For patient reported outcomes, the EDC will notify the study team if assessments are overdue so the study team can contact the participant up to three times to obtain these data. If a participant reported outcome is not completed within the scheduled window, the data will be considered missing with reasons provided. Missing data will be reported with reasons given where available and imputed using multiple imputation by chained equations (MICE) for the ITT population if more than 10% of TFI change scores cannot be determined due to missing data. The imputed data will be used in a sensitivity analysis to assess the impact of missing data on the primary endpoint.

Health economic evaluation.

A health economic evaluation forms an integral part of the trial and will be conducted from the perspectives of the NHS, individuals and society. The economic evaluation will take the form of a cost-utility analysis with a one-year time horizon using quality-adjusted life years (QALYs) as the main outcome measure. Although the clinical study has a non-inferiority design and our hypothesis is that QALYs will be similar between the Oto app and standard care, there is no prior hypothesis that costs will be similar between groups. Therefore, a standard economic evaluation framework will be used to synthesise differences in costs and QALYs between study arms using the incremental cost-effectiveness ratio (ICER) [25].

We will follow all participants from baseline to 12 months and will collect information about the delivery of the intervention (app and standard care), primary and community health care, visits to emergency departments and prescribed medications. We will focus on the collection of resource utilisation related to tinnitus. Data will be collected using bespoke questionnaires that will be embedded into the Lindus Health EDC platform for participant completion at regular intervals during follow-up (weeks 12, 26 and 52). Unit costs from recognised national sources including the new NHS National Cost Collection, the Personal and Social Services Research Unit (PSSRU), and the NHS Electronic Drug Tariff will be used to value NHS health care resource use. To understand the burden beyond the health service, we will ask participants about any related out-of-pocket expenses (e.g. hearing aids, complementary and alternative paid treatments) and any time away from paid employment due to the condition.

Tinnitus is known to impair aspects of daily health-related quality of life (HRQoL) and the EQ-5D-5L and the HUI3 will be administered at baseline, and 12, 26 and 52 weeks from randomisation. The health state described in each measure will be converted into a single value on a scale where death is anchored at 0 and perfect health at 1 using value sets. For the EQ-5D-5L the interim 5L- > 3L mapping tool will be used to estimate utility values [26]. To calculate utility values from HUI3 responses the Canadian tariff will be used as a UK-specific value set is currently not available. The questionnaires will be administered at baseline and at all follow-up points (weeks 12, 26 and 52). QALYs for each participant over the trial period will be calculated and will be derived by estimating the area under the curve of the profile generated connecting estimates of HRQoL at each time point using linear interpolation. If necessary, any baseline imbalances in HRQoL at baseline and resulting QALYs will be adjusted using regression methods. QALYs will be derived from utility values from the EQ-5D-5L and HUI3 but the former will be used in the base case analysis.

To understand the burden for participants and wider society, healthcare costs will be reported separately from any out-of-pocket expenses and any societal productivity losses. When analysing continuous variables (for example costs, HRQoL scores, QALYs) means and standard deviations will be computed for each trial arm and for comparisons between trial arms, mean differences and 95% parametric confidence intervals will be used. Mean differences in costs and QALYs between study arms will be synthesised using ICERs and to facilitate interpretation, net benefit statistics will be calculated [27]. Uncertainty around net-benefits will be expressed using confidence intervals (if feasible) and cost-effectiveness acceptability curves [28]. Deterministic sensitivity analysis will be used to address uncertainty around study parameters (e.g. unit costs) not subject to sampling variation. Data completeness on resource use and HRQoL will be closely monitored on a regular basis and missing data mechanisms discussed. Multiple imputation will be implemented to handle missing data if a complete case analysis results in substantial loss of information that could bias our results [29]. All analyses will be conducted and reported in accordance with existing good practice guidelines for economic evaluation [30, 31].

Sub-study sampling.

The DEFINE Trial Manager will generate a list of participants who have consented to further contact for the focus group. Targeted sampling will be conducted 4 months after the start of trial enrolment, and then further rounds will be undertaken 7 and 13 months after first enrolment. Participants will not be selected for more than one focus group timepoint. Where possible the selected sample will be weighted to Office of National Statistics criteria for a nationally representative group based on several variables (age, sex, ethnicity, geographical region), balanced across the randomised intervention given, and the severity of tinnitus (baseline TFI). Consent for focus group participation and data analysis is via a separate consent form for the sub-study.

Interview and focus group design.

Focus groups and individual participant interviews will be semi-structured, guided by a topic outline. The first set of interviews will be conducted at approximately 12 weeks following allocation, consisting of 1:1 interviews with approximately 10 participants. The second round of participant engagement will occur at 26 weeks following allocation, featuring focus group discussions with 4–5 participants in each group. Focus groups and interviews will be conducted online using video conferencing software. The selection of participants will ensure diversity in age, sex, ethnicity, geographical location, and tinnitus severity. A trained facilitator who is part of the DEFINE trial team will coordinate and guide both the individual and group interviews.

Focus group topics.

A semi-structured topic guide will be used by the facilitator to initially engage participants in key topics and prompt focussed discussions (S5 File). The guide will explore: barriers to seeking and obtaining advice and treatment for tinnitus; participant ideas, concerns and expectations about the trial interventions; participant interaction with the trial interventions; potential improvements to the delivery or content of the trial interventions; deployment of the interventions within the NHS.

Analysis and outcomes.

Transcript files will be downloaded from Microsoft Teams and checked against the original video recording for accuracy. Transcripts will be analysed using an inductive thematic analysis, following a 6-step process [32] adapted for group analysis [33]. The initial familiarisation phase will involve two investigators jointly reviewing the focus group data to discuss and establish an overall sense of the data. The investigators will then read the transcripts individually before working jointly again to generate codes and themes. The themes will then be reviewed and defined, before producing the report.