German adaptation of the Resources for Enhancing Alzheimer’s Caregiver Health II: study protocol of a single-centred, randomised controlled trial

The study has been designed as a non-blinded, randomised, and controlled trial. The intervention group receives the DeREACH intervention and the control group receives the standard supply of health-care services; both groups receive baseline, final, and follow-up assessments.

The design, conduct, and reporting of the DeREACH study will adhere to the Consolidation Standards of Reporting Trials (CONSORT) guidelines [7].

We only include informal caregivers of people with dementia who live at home with their relatives and are residents of Leipzig. In addition, participants must be able to speak German fluently and a diagnosis of dementia should have been established by a physician for each care recipient.

The recruitment process is supported by the Geriatric and Gerontopsychiatric Network of Leipzig. This network was established to advance the collaboration of various health-care systems and to assure that people with dementia in particular receive constant attention. It involves memory clinics, home-care agencies, nursing homes, voluntary helpers, associations, and day-care facilitates. These participating facilities suggest eligible subjects, inviting them to attend the study. Furthermore, we address potential participants via articles and announcements in newspapers and via posters and flyers that are available in several health-care facilities.

The participants are split into the study groups randomly. To obtain the same number of subjects in the intervention and the control group, we use a randomisation in blocks [8]. Moreover, we use a central randomisation via randomisation lists realised by an online procedure of the Medical Faculty of the Ludwig-Maximilians-University of Munich [9]. The outcome rater is not informed about the randomisation code of a participant before opening a sealed envelope with the code inside after completion of the baseline assessment.

The informal caregivers, their relatives with dementia, and the study staff cannot be blinded in matters related to the study group.

A nurse from University Hospital Leipzig makes contact with our participants and arranges the first appointment. One or two outcome raters visit informal caregivers at home to conduct the baseline assessment. In addition, we undertake a dementia screening for diagnosing the degree of cognitive impairment [10]. This takes place during the first appointment at home with the diseased individual to confirm the diagnosis of dementia.

After completion of the baseline assessment, an envelope containing the subject’s randomisation code is opened to allocate the participant to the intervention group or control group.

Study staff will make another appointment for the first home visit with the informal caregivers allocated to the intervention group. Altogether, the informal caregivers in the intervention group receive nine home visits and three phone sessions. Every home visit is structured in its sequence and appointments will be scheduled visit by visit. Usually, the whole intervention lasts six months and a post-intervention assessment takes place immediately after completion. A follow-up assessment takes place nine months after the baseline assessment.

The informal caregivers of the control group only receive the standard supply of health care services and the assessments outlined above. An overview of the study design and the assessment points is provided in Figure 1.

The intervention is executed by medical staff (for instance, nurses and occupational therapists). Prior to the intervention, these interventionists obtain training and a manual related to the intervention. Furthermore, they are attended and supported by research staff.

For this trial, the original REACH II intervention will be translated into the German language. Regional distinctions have to be adapted to meet conditions in Germany. Especially with regard to the areas of social support and self-care, it was necessary to include regional facilities and other offers of support. All 12 sessions of the DeREACH intervention are described in a study manual. Departing from the original REACH II intervention, we will train medical staff and not volunteers to perform the intervention. Another feature that required modification was the computer telephone integration system that the REACH II intervention provided to all participants, because the requirements are not available in the study area.

DeREACH is a multi-component intervention involving various strategies and techniques, such as psychoeducational information, problem-solving tasks, risk management, role playing, stress management, and social support. Five potential areas of risk in care giving receive particular attention in the intervention: safety, emotional well-being and self-care, social support, problem behaviour and preventive health behaviour. The intervention occurs over six months and includes 12 individual sessions (nine at home and three by telephone). Table 1 outlines a typical home session. A detailed description of the original REACH II intervention is available at the Epidemiology Data Center of the University of Pittsburgh [11].

Demographic information from the informal caregivers as well as from the people with dementia will be gathered via a questionnaire during the baseline assessment. In this trial, the outcome of primary interest is the effect on the burden of care-giving. Caregiver burden is perceived as a multi-dimensional response to stressors associated with care giving. We use a translated version of the Zarit Burden Inventory (ZBI [12]). The ZBI is a widely used 22-item assessment tool for measuring the caregiver’s perceived burden [13, 14]. The data will be gathered at baseline, after intervention completion (six months), and at the follow-up nine months after baseline. Secondary outcomes are changes in the depression [15], somatisation [16], health-related quality of life [17] and social support of the family caregivers [18]. In addition, with regard to the relative with dementia, two further secondary outcomes will be collected: the effects on health-care service utilisation and problem behaviour [19] (see Table 2).

The calculation of sample size was based on an updated meta-analysis determining the effectiveness of interventions for family caregivers of older adults [20]. In that study, a moderate mean effect size of 0.65 (95% CI [0.46-0.84,]) was determined for multi-component interventions on caregiver burden. Assuming this effect size as significant for DeREACH, which is a multi-component intervention, and using a t-test for two independent samples with a 0.05 two-sided significance level, a sample size of 51 patients per group was calculated to have 90% power for detecting a treatment effect of that size (calculated by G*Power [21]). Allowing for a drop-out rate of 35%, a total of 69 informal caregivers per group is planned.

All analyses will be conducted using Statistical Packages for Social Sciences (SPSS) version 20.0 for Microsoft Windows. The change from baseline to post-intervention assessment (six months) on the ZBI (primary outcome) will be compared between treatment groups using t-tests for independent samples on all randomised informal caregivers (intention-to-treat population). In case of significant between-group differences at baseline, an analysis of the covariance will be used with the post-intervention score as a dependent variable and treatment group and baseline ZBI score will be used as additional covariates. Several SPSS methods with missing value imputation will be used, including expectation maximisation (EM) and multiple imputations. Secondary analyses of the primary outcome variable will be performed on informal caregivers treated per protocol (PP population). Secondary study outcomes will be subjected to an explorative analysis. Post-hoc analyses will be performed on the primary outcome variable in patient subgroups, defined for instance by age, sex, caregiver education, family relationship and cognitive ability of the care recipient at baseline. Secondary and post-hoc analyses will be explorative without multiplicity adjustment.

The study design, study protocol, procedure and informed consent are approved by the ethics committees of the Leipzig University (ref. 217-12-02072012). Participation is voluntary and all participants will sign informed consent.

Recruitment into the study started on 1 October 2012.