Study design

A cluster randomised controlled trial with an 18-month follow-up period will be performed. Cluster randomisation (with the HAD practice as a unit) was chosen over individual randomisation because the latter would hold a high risk of contamination. In case of individual randomisation, the HAD personnel would have to switch between approaches (SUPR/care as usual (CaU)) frequently and could accidentally refer to or offer SUPR to clients assigned to the CaU group. In addition, as the time between informing the clients about the study, receiving clients’ consent and the start of SUPR/CaU was relatively short, performing randomisation on an individual level was not feasible. Dutch HAD practices and consequently all clients in these practices were randomly assigned to one of two groups. The control group received CaU which is hearing aid fitting only, while the intervention group received hearing aid fitting supplemented with SUPR.

Care as usual

CaU starts with a preparation appointment during which a screening pure-tone audiogram (only air conduction) is administered by the hearing aid dispenser. If the hearing loss in one or both ears is at least 35 decibel (dB) hearing level (HL) (averaged over the three frequencies 1, 2 and 4 kHz) in one or both ears, someone is considered potentially eligible for hearing aid fitting and more comprehensive audiometry is required. If the client is interested in hearing aids, his/her general wishes and goals are discussed after which the Amsterdam Inventory for Auditory Disability and Handicap (AIADH; Kramer et al)45 is handed out. Clients are asked to complete the AIADH at home and bring it along to the next appointment. The AIADH assesses hearing activity limitations and participation restrictions. Clients are asked to assign a CP and involve them throughout the rehabilitation (eg, bring them to appointments). During the next appointment, that is, the intake appointment, comprehensive audiometry (air and bone conduction, and speech audiometry) are performed. The results of all tests, the AIADH and the wishes of the client determine what type of hearing aid may be best suited for this person. The appropriate hearing aids will be selected and fitted directly (if available in the HAD practice) or in a subsequent fitting appointment. Fitting is followed by a trial period which usually lasts up to 4 weeks, during which people can try out the hearing aid and decide whether or not to purchase it. Depending on the client’s needs, fine-tuning or other follow-up appointments are scheduled. These can be scheduled during the trial period and after the device has been purchased.

Intervention: SUPR

SUPR consists of a Practical Support Booklet and online elements. In addition, clients are asked to assign a CP who is involved actively in the programme (see below).

Practical Support Booklet

The Practical Support Booklet will be handed out at the end of the preparation appointment (first-time HAUs, experienced HAUs) or the intake appointment (experienced HAUs). The aims of the Practical Support Booklet are to: 1) assist clients and CPs in getting familiar with their hearing aid, 2) stimulate clients’ use of the hearing aid and clients’ and CPs’ use of communication strategies and 3) guide clients and their CPs through the various stages (ie, appointments) of the rehabilitation trajectory. Although the theoretical elements of the booklet can also be used as a reference after the purchase of the hearing aid, the booklet’s focus is on the period between the first HAD appointment and the end of the trial period. The booklet covers four parts, corresponding to the four key appointments during the trial period (ie, preparation appointment, intake appointment, control-tuning and/or fine-tuning appointment and purchase appointment). The information that is provided is synchronised with the topics which are typically discussed during these appointments. The first part outlines the process of getting a hearing aid and includes an introduction to the hearing aid dispenser’s care and an explanation about the pure tone audiogram. The client is asked to write down and rank specific communication goals (s)he wishes to reach by the end of the trial period (eg, "I want to be able to hear the stories of my 10 year old granddaughter Anne when I pick her up from school every Monday". The second part revolves around the types of hearing aids available and the client’s hearing aid preferences. Information about how to operate and maintain the device is provided as well. In the third part, the client and the CP are asked to write down their experiences with the new hearing aid and its settings. This information will be used for further refinement of the fitting. The final section of the booklet provides information on assistive listening devices, reimbursement of costs, more information on the audiogram, types of hearing loss and the types of hearing aids (eg, behind-the-ear, in-the-canal or receiver-in-the-ear). In addition, an overview of the most important communication strategies that clients and their CP can apply is provided. The content and the appearance of the booklet were developed over the course of several months by the HAD company. Although no specific guidelines were used for the development of the written health information in the booklet, a number of the subsequent steps that are deemed important by Caposecco et al were taken into consideration: 1) interviews with key stakeholders (clients, CPs, HAD practice personnel) were held to specify the booklet’s goals and functions, 2) graphics and text were developed and optimised with regard to their understandability and attractiveness (language difficulty, layout, font size, paragraphing), 3) a first complete version of the booklet was pilot-tested in 10 HAD practices for several months. Feedback by all key stakeholders was collected and 4) the feedback was incorporated in a new and final version of the booklet (which was used in the study).46

Online elements

After the intake appointment, the links to the online elements will be sent to the participants via email. There are 2 emails which offer contact with the HAD practice and 11 emails which contain the links to the various educational videos that are offered (see below). The online part spans a period of up to about 6 months after the hearing aid purchase. The exact duration of SUPR depends on the duration of the trial period. For example, if a trial period is finalised in 3 weeks instead of the average four, the total duration of SUPR is 1 week shorter.

The educational videos consist of: 1) training modules on hearing aid handling skills. These comprise three short instruction videos with practical information on the use and maintenance of hearing aids. Participants receive the link to the relevant instruction video depending on their style of hearing aids (ie, behind-the-ear, in-the-canal or receiver-in-the-ear); 2) training modules on communication strategies and personal adjustment. This is a remake (ie, a modernised version) of the home educational programme ‘Horen en Gehoord Worden: Hoe kan het beter’, as developed by Kramer et al.37 It comprises five short videos showing the difficulties that hearing-impaired people can experience in everyday listening situations. The typical reactions by both the hearing-impaired people and his/her social environment to these situations are shown, and a trainer illustrates how communication could be improved by using communication strategies (for both hearing-impaired people and his/her CP); 3) three testimonials by hearing-impaired peers who share their experiences with hearing aids.

Measurements

For all participants four measurements will beconducted: at baseline (after the preparation appointment, but before the actual hearing aid fitting) (T0), 6 months after the hearing aid purchase (T1), 1 year after the hearing aid purchase (T2) and 18 months after the hearing aid purchase (T3). Measurements at T3 serve to determine the long-term effects of SUPR, that is, 1 year after its completion. Data will be collected using online questionnaires through Survalyzer, which is an online survey programme. Email reminders will be sent within a week after the first invitation-email and another week after the first reminder, if necessary.

Study population and recruitment

The following procedures were followed during the recruitment period (February 2016 to September 2016). Hearing aid dispensers invited clients to participate in the study. First-time HAUs were invited at the end of their preparation appointment. Experienced HAUs were invited at the end of their preparation or at the end of their intake appointment, if they did not require a preparation appointment. Hearing aid dispensers handed out an information package including an invitation letter, a selection form outlining the inclusion and exclusion criteria, a brochure about the study and an envelope with an information letter and brochure for the CP. All interested participants were asked to enrol themselves for the study by subscribing on a registration webpage and signing the online consent from there. Every month, the number of clients who were invited (number of envelopes that was handed out) and were enrolled (number of online subscriptions) per HAD practice were determined. When enrolment numbers for a particular HAD practice were relatively low, a phone call was made to the specific HAD practice to notify them of their current number of enrolments, to identify possible underlying reasons, and to motivate them to reach the required target. Throughout the recruitment period, the HAD headquarters organised motivational conference calls for the HAD practices that had not yet reached their target. Finally, when enrolment ratings continued to be behind target, employees of the headquarters directly invited potentially eligible clients who were not invited by the HAD practice personnel, via a telephone call. The information package was then sent via mail.

Incentives

After completing the T0 questionnaire, all participants will be offered a voucher of €50 to spend on a hearing aid or €25 to spend on other articles of the HAD practice if they decide not to purchase a hearing aid. CPs will be offered a gift card. In addition, participants in the control group will be offered a shortened version of SUPR after 18 months. For them, SUPR will be slightly adjusted such that it became suitable for individuals who had already started using a hearing aid.

Employees of the HAD practices will beoffered gift cards once the total number of participants is recruited (see ‘Sample size calculation’ section).

Inclusion criteria

The following inclusion criteria for the hearing aid candidates were applied:

1) age 50 years or older; 2) willing to try out one or two new hearing aid(s) (ie, agreed to plan a follow-up appointment). This hearing aid could be their first (ie, first-time HAUs) or a replacement hearing aid (ie, experienced HAUs). Clients who did not purchase a hearing aid after the trial period were considered dropouts; 3) sufficient understanding of the Dutch language; 4) access to a personal computer with internet access and owner of an email account for the total duration of the study.

Exclusion criteria

The following hearing aid candidates were excluded: 1) candidates who received additional care at a specialised Audiology Clinic. In the Netherlands, an Audiology Clinic offers elaborate, multidisciplinary and specialised, tertiary healthcare and is aimed at people with complex hearing problems. This care may overlap and/or interfere with that of SUPR; 2) candidates who received a hearing aid primarily to suppress tinnitus complaints. For these individuals, the focus of the rehabilitation is not on restoring communication per se, and as such, they were not part of the target group of SUPR.

Although all participants were encouraged to assign a CP, it was not obligatory for them to assign one in order to participate in the study. For the CPs, the only inclusion criterion applied was that they should be 18 years or older.

Outcome measures

An overview of all outcome measures and measurements over time according to Standard Protocol Items; Recommendations for Interventional Trials is attached (see online supplementary appendix 1).47

Primary outcome measures—HAUs

• The use of communication strategies will be measured using the reliable and validated Dutch 35-item version of the Communication Profile for the Hearing Impaired (CPHI).48 49 Communication strategies are measured using the following subscales: maladaptive behaviours, verbal strategies and non-verbal strategies. Each subscale consists of statements for which the respondent has to indicate how often (s)he applies this strategy. An example: "I avoid conversations with strangers, because of my hearing loss" (subscale maladaptive behaviour). The five response options range from ‘almost never’ to ‘almost always’. Scores are averaged per subscale and range from 1 to 5. Some items were recoded because of reverse scaling. High scores indicate favourable strategies, whereas low scores indicate unfavourable strategies.

We have chosen for the communication strategies subscales of the CPHI as central outcome measures for the following reasons. First, the subscales are purported to measure the constructs that are acted on by the core active element of the intervention (ie, the revised home education programme). Second, the CPHI has proven to have very good validity and reliability in the target population of this study.49

Secondary outcome measures—HAUs

• Personal adjustment to hearing impairment will also be measured using the reliable and validated Dutch 35-item version of the CPHI.48 49 This second section of the CPHI deals with personal adjustment and also contains three subscales: self-acceptance, acceptance of loss and stress and withdrawal. An example item of the latter subscale is: "I feel very tense because of my hearing loss". The five response options range from ‘totally disagree’ (one) to ‘totally agree’ (five). All items were recoded because of reverse scaling. After recoding the item scores, average scores per subscale can be calculated, with low scores indicating poor personal adjustment and high scores indicating good personal adjustment.

• Self-efficacy of hearing aid handling will be measured by the basic handling subscale of the Measure of Audiologic Rehabilitation Self-Efficacy for Hearing Aids (MARS-HA). The English version of this 7-item subscale has good psychometric quality.50 Scores can range from 0% to 100%, with lower scores representing less certainty in one’s capability of handling a hearing aid. At T1, T2 and T3, the 5-item subscale advanced handling will be additionally administered. Dutch versions of the scales were created using the forward-backward method.51 At T0 ‘expected self-efficacy’ will be administered, whereas at T1, T2 and T3 ‘experienced self-efficacy’ will be determined as the new hearing aids will have been fitted by then. For measurement of ‘expected self-efficacy’, all MARS-HA-items start with "I think I can …", whereas for measurement of ‘experienced self-efficacy  all items start with "I can…".

• Hearing aid use. Self-reported use will be measured using the first item of the International Outcome Inventory—Hearing Aids (IOI-HA) ("How many hours per day on average have you been using your hearing aid(s) in the last two weeks?"). Response options are ‘none’, ‘less than 1 hour a day’, ‘1–4 hours a day’, '4–8 hours a day’ and ‘more than 8 hours a day’.52 Self-reported hearing aid use will additionally be measured by three questions from the use questionnaire developed by Laplante-Lévesque et al.53 The latter questionnaire was translated into Dutch, using the forward-backward method.51 Hearing aid use will also be measured objectively via data-logging.

• Self-reported intervention outcomes (hearing aid rehabilitation and SUPR outcome). IOI-HA (items 2–7) and the equivalent IOI-AI (all seven items) will be used to assess the outcome of hearing aid rehabilitation and SUPR, respectively.52 The Dutch version of IOI-HA has a good test–retest reliability and validity.54 The first item of the IOI-AI determines the frequency of the use of the alternative intervention, that is, "How often have you used the learnt communication strategies on an average day in the last 2 weeks?" Response options are ‘never’ (1), ‘rarely’ (2), ‘sometimes’ (3), ‘often’ (4) and ‘almost always’ (5). Items 2–7 of the IOI-HA/IOI-AI questionnaire cover: benefit, residual activity limitations, satisfaction with the hearing aid(s)/SUPR, residual participation restrictions, impact on others and quality of life.

• Satisfaction with the HAD practice service. Satisfaction will be measured by the following question: "How likely is it that you would recommend the service of the HAD practice to other people (family, friends, colleagues)?" It is scored on a visual analogue scale running from 0 (=not at all likely) to 10 (=extremely likely).

• Self-reported activity limitations and participation restrictions are measured using the reliable and validated original (Dutch) version of the AIADH.45 55 It contains 28 questions regarding everyday listening situations. An example is: "Do you immediately look into the right direction when somebody calls you in the street?" The 4-point response scale covers: ‘almost never’ (1), ‘sometimes’ (2), ‘often’ (3) and ‘almost always’ (4). When the participant answers the question with ‘almost never’ or ‘sometimes’, he or she is directed to a next question, which is about the inconvenience of not being able to hear well in that specific situation. Response options are: ‘no’ (1), ‘a little’ (2), ‘very handicapped’ (3) and ‘extremely handicapped’ (4). Hence, the total score can range from 28 to 112 with higher scores indicating greater participation restrictions.

• Readiness to do something about one’s hearing problems will be measured by the validated Dutch 24-item version of the University of Rhode Island Change Assessment.56 Formulations of items were adjusted such that they applied to hearing problems. The inventory contains 24 statements regarding attitudes and behaviours assessing an individual’s stage of behaviour change. At T0, the scores on the following stages will be assessed: pre-contemplation (does not intend to take action in the foreseeable future, eg, "As far as I’m concerned, I don’t have any problems with my hearing that need changing"), contemplation (intends to change in the next 6 months and is aware of the pros and cons of changing), and action (has made specific modifications in his/her lifestyle towards healthy behaviour). At T1, T2 and T3, the maintenance stage (can maintain the changes in new behaviour) will be added. The five response options range from ‘fully disagree’ (score one) to ‘fully agree’ (score five). Summed scores for each subscale will be calculated. In addition, the composite ‘readiness score’ (adding the contemplation, action and maintenance scores and subtracting the precontemplation score) and the composite ‘committed action score’ (subtracting the contemplation stage score from the action stage score) will be calculated.56 The higher the composite scores, the further the respondents are along the stages of change.

• Emotional response to hearing problems. The Hearing Handicap and Disability Inventory will be used.57 The purpose of the inventory is to identify the individual’s problems caused by hearing loss. Only the section ‘emotional response’ will be administered. It contains five statements each with five response options: ‘yes!’ (4), ‘yes’ (3), ‘more or less’ (2), ‘no’ (1) and ‘no!’ (0). An example is: "I find it difficult to accept that I am hearing impaired". Lower scores indicate better outcomes.

Secondary outcome measures—CP

• Third-party disability will be measured using the Significant Other Scale for Hearing Disability.12 This questionnaire was translated into Dutch for the purposes of this study following a forward-backward method.51 The 27-item questionnaire addresses the problems and limitations experienced by the CP. An example item is: "Because of my partner’s hearing difficulties I have to repeat myself often". For each item, the CP has to indicate how much of a problem it is for him/her: ‘no problem’ (0), ‘a mild problem’ (1), ‘a moderate problem’ (2), ‘a severe problem’ (3), ‘a complete problem’ (4). Higher scores indicate greater difficulties.

• Hearing aid rehabilitation and SUPR outcome as viewed from the perspective of the CP will be administered with the 7-item IOI-HA-SO/IOI-AI-SO and covers use, benefit, residual activity limitations, satisfaction, residual participation restrictions, impact on others and quality of life.58

Baseline measurement—demographical characteristics

• Gender (male/female).

• Age (in years).

• Marital status (married/cohabiting/widow or widower/divorced/single, never married).

• Living situation (living together with my partner/living together with my partner and children/living together without my partner but with one or more family members/living alone (own room) or in a care institution/living alone, independently or nursing home/other, namely…).

• Level of education (no completed education/lower general education, elementary education or a part of it/lower general secondary education/vocational education/secondary education/technical and vocational education/higher professional education/higher general education/scientific education/other, namely…).

• Occupational status (yes/no).

• Country of birth (The Netherlands/other, namely…).

• Country of birth father (The Netherlands/other, namely…).

• Country of birth mother (The Netherlands/other, namely…).

• Hearing loss in each ear, in dB HL (averaged over 1, 2 and 4 kHz) as retrieved from the pure-tone audiogram as provided by the hearing aid dispenser.

Randomisation

HAD practices were randomly assigned to offer CaU or the intervention. To avoid an unequal distribution of HAD practices with regard to level of urbanisation, HAD practices were prestratified (HAD practices located in a relatively rural area vs in an urban area) and randomisation occurred within these two strata. A statistician performed block randomisation of the HAD practices in the statistical software R, with random permutation in blocks of size four and with a fixed seed. Thirty-four HAD practices were assigned to CaU and 36 HAD practices to the intervention group. The recruitment procedure and period was the same for all 70 included HAD practices (the total list of included HAD practices are available on request from the research team).

Sample size calculation

Sample size calculations are based on the expected effects of the intervention on the primary outcomes: communication strategies (CPHI). Demorest and Erdman indicated that the expected difference on the subscales of the CPHI varies from 0.67 (maladaptive behaviour) to 0.95 (self-acceptance).59 Given that in a previous study,37 the effect of the programme was larger for first-time than for experienced users, we calculated sample sizes separately for first-time and experienced users. For first-time HAUs, we based our sample size calculations on an expected difference of 0.67 between the intervention and the CaU group. Note that the subscale with the smallest minimal importance difference (ie, maladaptive behaviour) was used in the calculation, as finding a significant difference on this measure requires the largest number of participants. Calculations in PASS 12 (tests for two means in a cluster-randomised design; intracluster correlation coefficient: 0.01; alpha: 0.05; power: 0.80) shows that when 70 HAD practices are included (of which half will offer SUPR and half will offer CaU), the number of first-time HAUs to include in the analyses is two per HAD practice. For the sample size calculation of the experienced users, we chose an expected difference of 0.4 between the intervention and CaU group. The expected difference was set lower than for first-time HAUs as Kramer et al had previously found generally smaller effects for experienced users than for first-time users.37 With a difference of 0.4, the number of experienced HAUs (power: 0.80) to include is three per HAD practice. We expected the proportion of dropout or loss to follow-up across the study to be 20%. This includes loss to follow-up for a range of reasons: no motivation anymore, reluctance to purchase a hearing aid after a successful trial, sickness, death, etc. Taking the loss to follow-up and the proportion of clients that normally purchase a hearing aid into account results in a total (rounded) number of four first-time HAUs per HAD practice and five experienced HAUs per HAD practice to be recruited.

Statistical analyses

To check the comparability between the groups (CaU or intervention group) at baseline, baseline characteristics of the participants will be compared using the χ² test (for categorical variables), the independent samples t-test (for normally distributed continuous variables) and the Mann-Whitney U test (for non-normally distributed continuous variables). Comparability will be checked for all demographic variables and all primary and secondary outcomes.

For the effect analyses, the groups will be compared on all primary and secondary outcome measures using linear mixed models including the results at T0, T1, T2 and T3. Group, time and their two-way interaction will be included as fixed effects in the mixed models, with random intercepts for subject and HAD practice. For the covariance matrix, a variance component structure will be chosen. To adjust for potential bias associated with multiplicity of analyses, statistical significance levels will beset at p<0.016 (0.05/3). If a significant effect will be found, an independent samples t-test was used and a Bonferroni correction will be administered because of multiple comparisons. Type of HAU (first-time or experienced) will be tested as an effect modifier for potential subgroup differences.

In case of substantial missing data, multiple imputation will be applied. The main analysis was intention to treat. Any outcome measure to be collected for participants who discontinue or deviate from intervention protocols will be saved and analysed according to the intention-to-treat protocol. In addition, a per-protocol analysis will be performed. A per-protocol analysis includes those participants who completed the intervention originally allocated as described in the study protocol. As a per-protocol analysis can potentially yield biassed effects (eg, see CONSORT statement),60 great caution will be exerted when interpreting these results. In addition, the report of these findings in future articles will be nuanced explicitly and thoroughly.

Process evaluation

The process of implementing SUPR into the HAD care in the intervention arm will be evaluated. The main aims of this evaluation are to gain insight into 1) the circumstances in which the intervention was implemented, 2) (non-) compliance with the intervention and 3) the professionals’ and clients’ appraisal of the intervention.

The process evaluation will be carried out according to the framework as proposed by Linnan and Steckler.61 It covers seven parameters: recruitment, reach, fidelity, dose delivered, dose received and implemented, satisfaction and perceived benefit.62 A brief description of each of the parameters is given below.

• Recruitment refers to the procedures applied to approach and attract potential participants. The hearing aid dispensers will be asked to provide this information.

• Reach. This is the proportion of people participating relative to the number of people invited.

• Fidelity relates to the question of whether the intervention was provided as intended. The team that is responsible for the email contact will be asked to provide a written report on this.

• Dose delivered: 1) Did the personnel of the HAD practice hand out the Practical Support Booklet at the end of the preparation appointment? 2) Did the personnel of the HAD headquarters send out the emails correctly (correct content) and on time?

• Dose received and implemented: 1) Did the participants receive and use the Practical Support Booklet? 2) Did the participants open the emails and the videos? If so, did they watch the whole video, or part(s) of it? The video watching behaviour will be determined using Quadia (supplier of online video content) and Google analytics. Data on the average watching time per video, and how many times a particular video has been opened will be determined. Due to the privacy regulations the HAD company is subject to, the company is only allowed to collect video watching data on a group level (and not on an individual level). As all the HAD practices of the company that do not participate in the study provide SUPR as their standard care at the time of the study, the researchers will not be able to determine specific group averages of the study participants (the averages are based on both study participants and regular HAD clients). Information on implementation of the knowledge that participants learnt from SUPR will be deduced from the IOI-AI questionnaire (item on use) on T1. If participants received and used the Practical Support Booklet will be measured by a questionnaire.

• Satisfaction: satisfaction of the participant with SUPR will be evaluated using the IOI-AI questionnaire (item satisfaction) on T1. The hearing aid dispensers will be asked to answer the question: How would you rate your satisfaction with SUPR?

• Benefit: information on the experienced benefit of the participant will be obtained from the IOI-AI questionnaire (item benefit) on T1. The hearing aid dispensers will be asked to answer the question: How would you rate the perceived benefit from SUPR for your clients’ ability to improve in communication?

Additionally, focus group discussions with participants from the intervention group will be organised to gain insight into the reasons for using the knowledge of SUPR in their daily lives or not. A minimum of two focus groups will be organised. The exact number will depend on data saturation. Heterogeneity in age, gender, educational level, severity of hearing impairment and stage of behaviour change (at baseline) within the groups will bestrived for. Given the difficulties hearing-impaired individuals might have with group conversations, the focus groups will have a maximum size of six participants each.