The selection of measures was guided by previous research on both caregiver burden and suicidality and was informed by the experiences of a consumer advisory panel comprising peak bodies for caring and dementia in Australia, service providers, and former carers. The survey comprised the following self-report scales:
The Revised Memory and Behavior Problems Checklist (Teri et al., 1992) is a 24-item measure of behavioural problems in the person with dementia and the caregiver's reaction to those problems. Two overall scores are created, where higher scores reflect, respectively, more frequent behaviours (RMBPC_Behavior) and more intense reactions (RMBPC_Reaction). The behaviour and reaction scales are reported to have reliabilities of 0.84 and 0.90, respectively (Teri et al., 1992). In the current study, Cronbach's alphas were 0.87 and 0.89, respectively.
The Fortinsky Caregiver Self-efficacy Scale (Fortinsky et al., 2002) is a 10-item measure of self-efficacy. Nine of the items are used to create two subscales—self-efficacy for symptom management (FSE_Symptoms) and self-efficacy for community support service use (FSE_Community). Higher scores indicate greater self-efficacy. Internal consistency for the two scales has been reported as 0.77 and 0.78, respectively (Fortinsky et al., 2002). In the current study, Cronbach's alphas were 0.86 and 0.82, respectively.
The Alzheimer's Disease Knowledge Scale (DKS; Carpenter et al., 2009) is a 30-item test of knowledge about Alzheimer's disease risk factors, symptoms and management. It comprises 30 statements, each with a true/false response option. For the purposes of this study, the term ‘Alzheimer's Disease’ in each statement was replaced with ‘dementia’. Correct responses are summed to create a total score, with higher scores indicating better knowledge. The reliability has been reported as 0.71 (Carpenter et al., 2009). In the current study, Cronbach's alpha was 0.51. Given the low alpha, the responses to each question were inspected. The proportion of participants answering correctly was more than 70% on all but four items (with the majority of items answered correctly by more than 90% of participants). The four items that were answered incorrectly by 30% of participants or more addressed life expectancy after diagnosis, blood pressure and cholesterol as risk factors, and the benefits of psychotherapy for depression in people with dementia.
The SF-12 Health Survey Version 2 (Ware et al., 2005) is a 12-item generic measure of health. For the purposes of this analysis, the Physical Health Component Summary Scores (SF12_PCS) was created according to the recommended algorithm (Ware et al., 2005). The Mental Health Component was not calculated because individual measures of specific psychological constructs (e.g. depression, hopelessness) were included in the survey. Reliability estimates for the Physical Component Summary have been reported as 0.89 (Ware et al., 2005), and in the current study Cronbach's alpha was 0.88.
The Center for Epidemiologic Studies Depression Scale (CES-D; Radloff, 1977) is a 20-item measure of depressive symptomatology in the previous week. Higher scores indicate more depressive symptoms. Internal consistency has been reported as 0.85 (Radloff, 1977). In the current study, Cronbach's alpha was 0.91.
The Beck Hopelessness Scale (BHS; Beck et al., 1974) is a 20-item scale comprising statements about attitudes towards the future. Higher scores indicate greater levels of hopelessness. The internal consistency has been reported as 0.93 (Beck et al., 1974). In the current study, Cronbach's alpha was 0.93.
The Geriatric Anxiety Inventory (GAI; Pachana et al., 2007) is a 20-item measure of anxiety symptoms in the past week. Higher scores indicate more symptoms of anxiety. The internal consistency has been reported as 0.91. In the current study, Cronbach's alpha was .93.
The Life Orientation Test (LOT; Scheier and Carver, 1985) is a 12-item measure of dispositional optimism. Eight items contribute to the final score, where higher scores indicate greater optimism. Internal consistency has been reported as 0.76 (Scheier and Carver, 1985). In the current study, Cronbach's alpha was 0.92.
The Zarit Burden Scale (Zarit et al., 1985) is a 22-item measure of the sense of burden resulting from providing care. In the current study, the Cronbach's alpha was 0.90.
The Brief COPE (Carver, 1997) is a 28-item measure of coping strategies. Guided by the work of Cooper et al. (2008), three subscales were created—emotion-focused, problem-focused, and dysfunctional coping strategies—with higher scores indicating greater use of the relevant strategies. Internal consistencies have been reported as 0.72, 0.84 and 0.75, respectively (Cooper et al., 2008). In the current study, Cronbach's alpha was 0.72, 0.82 and 0.64, respectively.
The Duke Social Support Index (Koenig et al., 1993) is a measure of social interactions and perceived support. The 11-item version, used in the Australian Longitudinal Study on Women's Health (Pachana et al., 2008) and comprising two subscales, was used. For the purposes of this analysis, only the satisfaction subscale (DSSI_Satisfaction) was used, with higher scores indicating a greater level of satisfaction with social support. Internal consistency has been reported as 0.80. In the current study, Cronbach's alpha was 0.88.
The Suicidal Behaviors Questionnaire-Revised (Osman et al., 2001) is a four-item measure of past suicidal thoughts and behaviours, and perceived likelihood of future suicidal behaviour. Item Two asks participants ‘How often have you thought about killing yourself in the past year?’, with response options of Never, Rarely (1 time), Sometimes (2 times), Often (3 to 4 times) and Very often (5 or more times). For the purposes of this analysis, participants who responded Never or Rarely were classified as ‘non-suicidal’, whereas those who responded Sometimes, Often or Very Often were classified as ‘suicidal’. Item Three asks ‘Have you ever told someone that you were going to commit suicide, or that you might do it?’, with response options of ‘No’, ‘Yes, at one time, but I did not really want to die’, ‘Yes, at one time, and I really wanted to die’, ‘Yes, more than once, but I did not want to do it’ and ‘Yes, more than once, and I really wanted to do it’. Participants who selected any of the ‘Yes’ responses were classified as having disclosed their suicidal thoughts or intentions. Item Four asks ‘How likely is it that you will attempt suicide someday?’, with response options of Never, No chance at all, Rather unlikely, Unlikely, Likely, Rather likely and Very likely. Participants who responded Likely, Rather likely or Very likely were classified as ‘likely to attempt’ with the remainder classified as ‘unlikely to attempt’.
A range of demographic and care-related information was also collected, including age, gender, relationship to person with dementia and length of time spent caring.