Study design and participants

This ongoing study is designed as a multi-site, two-arm RCT, with an intervention group and a wait-list control group (figure 1). The protocol is developed in accordance with the Standard Protocol Items: Recommendations for Interventional Trials recommendation.26 It is a 3-year study, and recruitment commenced in January 2017. This study aims to recruit participant pairs comprising a family caregiver (related to an individual diagnosed with AMD) and the care recipient (individual diagnosed with AMD). A family caregiver will be defined as ‘any person who, without being a professional or belonging to a social support network, and in some way, is directly implicated in the patient's eye care or is directly affected by the patient's health problem.’9

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Flowchart of randomised controlled trial. AMD, age-related macular degeneration; CBT, cognitive behavioural therapy.

Eligibility of participants

Recruitment

Participant pairs will be recruited primarily from private eye clinics in Westmead and Liverpool (Sydney, New South Wales, Australia). Recruitment involves an opt-in two-step process for a participant pair. Consent for participation is required from both the caregiver and care recipient. Step 1 involves screening for interest in participating in the research via either: (1) the study coordinator discussing the study with potential participants attending the eye clinic or (2) advertising the study through MDFA and Carers NSW newsletters and clients self-referring as a result of the provision of information. The advertisement will cover ‘possible emotional strain and stress’ associated with caring and the treatment's aims of ‘improving emotional well-being and coping.’

Step 2 involves providing further information about the study via the study information sheet and obtaining written consent for the participant by either: (1) providing the information sheets, consent forms and baseline questionnaires to potential participant pairs attending the eye clinic (this will be completed at the clinic or mail returned with a reply-paid envelope) or (2) telephoning the potential participant to further discuss the research, followed by posting information sheets, consent forms and baseline questionnaires to people who have indicated willingness to participate. Loss of contact will be minimised by asking for multiple contact methods.

Randomisation and blinding

All participants will be randomly assigned to one of two groups: (i) multi-modal treatment or (ii) wait-list control (figure 1). The randomisation sequence will be generated centrally using permuted blocks of mixed size, stratified by recruitment source (clinic versus self-referral as a result of responding to MDFA and Carers NSW advertisements), to ensure equal numbers while maintaining an unpredictable sequence. Assignments to the intervention or control group will be managed centrally, separately from recruiting and treating staff, in order to ensure that a randomisation number and group is assigned and recorded for each recruit before their allocation becomes apparent.

Intervention group

The intervention group will receive a multi-modal support service programme consisting of: (1) a brief M-CBT treatment to be delivered fortnightly as five individual modules and (2) five Talk-Link group counselling sessions (an existing Carers NSW support service). The Talk-Link Counselling and M-CBT will occur weekly on an alternating basis (ie, week 1: M-CBT, week 2: Talk Link, week 3: M-CBT, and so on). The whole intervention will be conducted over 10 weeks (table 1).

M-CBT of fortnightly modules (table 1) formatted in Microsoft PowerPoint, with additional homework worksheets and accompanying templates for practising acquired skills, will be mailed to participants in the intervention group. Each module will target a specific stressor and/or train a new adaptive coping method (table 1) and will be supported by targeted homework assignments for the caregiver to practice between sessions. These will draw on standard CBT practice and on a number of additional sources including published trials, empirical studies and treatment guides.17–20 Participants will be telephoned following each M-CBT module for a brief conversation (5–10 min) aimed at: (1) addressing any queries regarding the information presented in the most recent module, (2) encouraging the learning and practising of skills from the previous module, (3) normalising the challenges of acquiring new skills, (4) emphasising the importance of the consistent practice of newly acquired skills for positive impact and (5) reminding participants in the case of unread materials. We will also ask participants how much time they spent on each module that week including reading the materials and completing the related homework sheets. Calls will be logged for each participant. The educational materials sent out to participants will aim to increase their knowledge about support and respite services and financial entitlements and benefits including how best to access them, which has been previously shown to increase caregivers’ sense of competence and reduce depression.14

Participants in the intervention group will be asked to complete a form prior to participating in the Carers NSW Talk-Link group sessions to identify the issues they might want to discuss (table 1) and to also reinforce CBT skills taught in the prior week. These will then be developed into weekly topics to guide group discussions. Information/ newsletters will be provided by the facilitators. Using teleconferencing, a group of six to eight caregivers and two trained facilitators will converse over the telephone, at the same time each week, to explore issues around caring for someone with AMD. These sessions will last for 1 hour.

Control group

During the treatment phase, the active wait-list control group will receive reading materials concerning AMD and caring for persons diagnosed with the condition. All control participants will be offered the opportunity to receive the multi-component intervention after the study ends (12–18 months after inclusion).

Measurement of covariates

A preintervention baseline questionnaire will obtain demographic and health-related information (ie, covariate data) from all family caregivers: age, sex, years of education, employment status, living arrangements, marital status, whether they are the sole caregiver, relationship to care recipient and their own health status (eg, subjective self-rated health and any chronic conditions that they have). Follow-up assessments will be conducted by postal questionnaires for all participants (caregivers only and not care recipients) from both RCT arms: immediately after completion of the 10-week programme and 6 and 12 months after initiation of the intervention (figure 1).

General caregiving activities will be assessed by asking the caregiver whether they performed any of the basic (eg, toileting, feeding, dressing, grooming) and instrumental daily living activities (eg, housekeeping, food preparation, shopping, telephone usage and transportation) for their care recipient using a validated questionnaire.27 In relation to each of these activities, they will be asked to report whether they gave: (a) no help or little help given, (b) moderate amount of help given, (c) high amount of help given or (d) not applicable. For eye-related activities that the caregiver assisted with, questions will quantify the impact of the anti-vascular endothelial growth factor (VEGF) therapy received by the care recipient, if applicable: the number of appointments attended, the average time taken for appointments, time taken away from work and loss of income. Caregivers will be asked to detail the frequency (1 day/week; 2–3 days/week; 4–6 days/week; 7 days/week; 1–3 days/month or other) and duration of care (<1 hour; 1–2 hours; 3–5 hours; 5–8 hours; >8 hours or other) that they generally provide. The caregiver will also report the level of dependency of the care recipient on them, by choosing one of the following: (a) not at all dependent, (b) somewhat dependent, (c) moderately dependent, (d) very dependent or (e) extremely dependent.

The following information related to care recipients (patients with AMD) will be collected through the eye clinic records: visual acuity in the better eye, type of AMD (wet or dry AMD), number of anti-VEGF injections to date, number of clinic appointments and total follow-up period. A brief questionnaire will also be administered to care recipients at baseline, in order to collect information on socio-demographic details (age, sex, ethnicity, education, employment status, living arrangements and marital status), any other eye diseases (eg, cataract, glaucoma) and co-morbid chronic conditions they may have from, for example, heart disease, stroke, diabetes or arthritis. The questionnaire will also determine self-reported vision-related health status, using the National Eye Institute Visual Function Questionnaire (NEI-VFQ-25).28 This is a 25-item scale used to generate a single composite score from 0 to 100 (100 is maximum visual function).9

Outcome measurement

Outcome data will be obtained from all randomised study participants using validated questionnaires or tools conducted at baseline, 10 weeks and 6 and 12 months. Subjective caregiver burden is the primary outcome and will be measured using the Caregiver Burden Scale (CBS) (table 2). Secondary outcomes (ie, fatigue, depressive symptoms, health-related QOL (EQ-5D-5L) and self-efficacy) are also detailed in table 2. Furthermore, caregivers will indicate their overall QOL on a linear analogue uniscale from 0 (worst possible) to 10 (best possible). Caregivers will be asked to identify whether caring for a family member with AMD created any specific problems, including: feeling tired, anxious, stressed, sad and/or depressed; impediments to employment, retirement, social activities and travel plans; and/or relationship problems as a result of the caregiving experience.

Individual participant consent will be sought from caregivers for Medicare Benefits Scheme (MBS) data linkage. Permission will be sought to link data from 12 months prior to the intervention and up until 1 year after the intervention. Hence, health-related resource use data (eg, medical consultations and prescription use) will be compared between the two RCT arms and will contribute to the economic evaluation. To protect the participants’ privacy, all outcome data will be kept in a password-protected file that is separate from the document containing the participants’ names and identification codes. All documents related to the study will be stored in restricted-access facilities or locked cupboards and on restricted-access servers and password-protected personal computers and files as per University of Sydney human research ethics committee (HREC) requirements.

Acceptability and feasibility of the intervention

Treatment acceptability in the intervention arm will be assessed immediately after completion of the 10-week multi-modal treatment by asking participants to rate their satisfaction with the following: (1) overall support service programme they received, (2) CBT modules they completed and (3) adherence to the programme. Participants will respond to these three questions using a 5-point Likert Scale, ranging from ‘very satisfied’ to ‘very dissatisfied.’ Participants will also be asked to respond with yes/no to the following questions: (1) “Would you feel confident in recommending this treatment to a friend?”; (2) “Was it worth your time participating in this programme?” Finally, participants will be asked to specify which components of the programme they found most useful or valuable, and suggestions will be elicited on how the programme could be improved further. Moreover, the percentage of people who enter the study compared with the total number to whom participation is offered and the follow-up rate achieved will also be determined. Treatment feasibility will be broadly assessed based on preliminary outcomes achieved: cost-effectiveness of the intervention, combined with the ease with which support staff could support participants.

Economic evaluation

The analysis will be conducted from a societal perspective and will include costs incurred by individuals and healthcare providers. This perspective will be taken because the cost consequences of this intervention may extend beyond the domain of healthcare. This will be a ‘within-trial’ economic evaluation carried out within the timeframe and context of the trial and will determine the cost–utility of this support service compared with usual care. We will determine the cost per person to deliver the support service using a bottom-up approach. This will include the cost of training MDFA staff, telephone consults, educational materials and other consumables. These will be valued at market rates and presented in 2017 $A. This will enable us to determine the cost of wider roll out of the programme. Direct healthcare costs will be determined through patient-level linkage to MBS and  Pharmaceutical Benefits Scheme (PBS). Healthcare resource use during the trial will include (1) primary and secondary healthcare consultations and admissions and (2) use of medications, including antidepressants and anxiolytic medications. Healthcare costs will include both government contributions to healthcare plus any patient out-of-pocket expenses. We will also measure and value time contributions of caregivers to attend the course and take part in discussion groups, using median wage rates.

Participants’ health states will be captured using the five domains of the EQ-5D-5L, namely, mobility, self-care, usual activities, pain/discomfort and anxiety/depression. The responses on each domain of the EQ-5D-5L will be converted into utility values using a valuation algorithm for the Australian population.29Quality-adjusted life years (QALYs) will be determined over the duration of the trial from the summation of mean utility x time interval between EQ-5D administration times. The incremental cost-effectiveness ratio in $A/QALY will be determined in the intervention compared with control. Bootstrapping will be used to estimate a distribution of the joint uncertainty in mean costs and outcomes and to calculate the CIs around the incremental cost-effectiveness ratio. One-way sensitivity analysis will be conducted around key variables. A cost-effectiveness acceptability curve will be derived that summarises the probability of the intervention being cost-effective based on willingness to pay per QALY gained.

Sample size considerations

We plan to recruit 360 caregiver–patient pairs (180 in each RCT arm) over 4–6 months. Allowing 20% loss to follow-up due to changes in the status of care recipients or death/dropout of caregivers, we will have 288 caregivers available for analysis. For our primary endpoint (caregiver burden scores), 194 available caregivers provide 80% power to detect a moderate and clinically meaningful effect size of 0.5 SD of the intervention on caregiver burden, after allowing for full compliance of 81%, at α=0.05. An 81% full compliance level represents 90% of recruits in the intervention arm actively taking up the educational and support programme, among whom 90% actively practice the strategies learnt. We have chosen to recruit the higher number of 288 available caregivers, so that we will additionally be able to estimate rates of uptake, acceptance, satisfaction and improvement with intervention with high precision (standard errors <6%) and have well-powered secondary analyses including an effect size of 0.5 of intervention on mean overall QOL and a difference of 20% vs 40% between intervention and control groups in the proportion reporting specific negative impacts of caregiving.

Data collection and analysis plan

All data will be collected using questionnaire booklets and data collection sheets and will be subsequently entered into a secure online platform, called Research Electronic Data Capture. Primary and secondary study endpoints will be analysed by intention to treat. The primary study endpoint will be postintervention change from baseline in CBS scores. The primary analysis will test for difference between intervention and control groups in postintervention change from baseline using linear mixed models for repeated measures. Secondary analyses will directly test change in scores at each follow-up within the intervention and control groups and test for difference between intervention and control groups in the average of the three postintervention CBS scores. The percentage of individuals within intervention and control arms who demonstrate a large change in caregiver burden, defined as a change of half an SD or more, will be estimated with 95% CI. Secondary endpoints are: percentage of participants reporting overall satisfaction with the intervention and each major subcomponent, percentage of participants completing the overall programme and each major subcomponent, change from baseline in QOL, fatigue and self-efficacy scores and postintervention frequency of depressive symptoms.

Differences between the intervention and control groups in individual EQ-5D-5L domain and depression outcomes will be tested using χ2 tests, and mean postintervention level of caregiver burden, fatigue, self-efficacy, depressive symptoms and general health will be tested using two-sample t-tests. Rates of uptake, continued participation, acceptability and satisfaction with intervention will be estimated with 95% CI. Associations between baseline recipient and caregiver characteristics, treatment experience and postintervention outcomes will be explored using linear mixed models for repeated measures. Baseline characteristics of the intervention and control groups will be examined.

Safety measures and endpoint

A clinical psychologist (AC), experienced in the assessment and treatment of mental health disorders, is supporting this study. The occurrence of adverse events, such as clinical psychological presentation (ie, anxiety, depression) or risk of self-harm and suicide, is considered safety endpoint measures of this trial. Regular contact by support staff will identify whether the safety endpoint measure is required for any participant, and the clinical psychologist on the research team, as per accepted mental health clinical guidelines, would instigate appropriate action.

Ethics and dissemination

The study was approved by the University of Sydney human research ethics committee. The committee deemed this study as low risk in terms of ethical issues. The written papers from this study will be submitted for publication in quality peer-reviewed medical and health journals. Study findings will also be disseminated via presentations at local, national and international conferences.