Caregivers received general information on dementia care and follow-up phone calls simply to maintain contact, but without any training for developing a behavioral problem-management plan and strategies.

Caregivers received solutions for managing behavioral problems, with referrals to community services and telephone consultation, further assurance and consultation were provided in monthly telephone follow-ups, and progress in behavior management was evaluated.

A home-based caregiver-training program consisted of two weekly sessions, each lasting 2 to 3 hours. Following the training sessions, further assurance and consultation were provided in monthly telephone follow-ups, and progress in behavior management was evaluated.

The Agitation Management Self-efficacy Scale was used to measure caregivers' self-efficacy for managing dementia patients' agitation. Caregivers were asked how confident they were about handling the problem for each identified behavioral problem and if they believed that they could manage the problem for behaviors that did not occur. Scores range from 42 to 210, with higher scores representing greater caregiver self-efficacy. In this study, Cronbach's alphas ranged from 0.98 to 0.99 at different time points.

Preparedness was measured by the 10-item Caregiver Preparedness Scale,asks caregivers to rate how well prepared they think they are for seven domains of caregiving. A final question asks for an overall rating of how well prepared caregivers think they are to care for the care receiver. Items are scored on a 5-point Likert scale from 1 (not prepared) to 5 (well prepared). Scores range from 10 to 50, with higher scores representing greater preparedness for caregiving tasks. Validity and reliability of the original Preparedness scale was supported.24 The content validity index for the Preparedness Scale Taiwanese version was 1.0 and Cronbach's alpha for this scale among Taiwanese caregivers was 0.87. Cronbach's alpha in this study was 0.92.

A 17-item Competence Scale was used to assess caregivers' knowledge and skills for managing behavioral problems of patients with dementia. Scores range from 17 to 85, with higher scores representing better competence. In this study, Cronbach's alpha ranged from 0.90 to 0.93 at different time points.

Physically aggressive behaviors of dementia patients were measured by the PAB subscale of the Chinese version CMAI, community form, which was shown to be valid and reliable for a Taiwanese sample. Each item (behavioral problem) is scored according to its frequency from 1 (never happens) to 7 (several times per hour). PAB subscale scores range from 7 to 49, with higher scores indicating more physically aggressive behaviors. In this study, the PAB subscale had Cronbach's alpha of 0.55.

The Taiwan version of the Medical Outcomes SF-36 was used to measure family caregivers' HRQoL. The SF-36 contains eight generic health concepts: physical functioning (PF), role disability due to physical health problems (RP); bodily pain (BP); vitality (energy/fatigue) (VT); general health perceptions (GH); role disability due to emotional problems (RE); social functioning (SF); and general mental health (MH). Taiwan-specific SF-36 algorithms were used to compute the Mental Component Summary (MCS) and Physical Component Summary (PCS) scores using norm-based (mean = 50, SD = 10) scoring methods. Scores for each scale range from 0 to 100, with higher scores representing better health outcomes. In this study, Cronbach's alphas for the eight scales ranged from 0.81 to 0.99.

Caregivers received general information on dementia care and follow-up phone calls simply to maintain contact, but without any training for developing a behavioral problem-management plan and strategies.

The Agitation Management Self-efficacy Scale was used to measure caregivers' self-efficacy for managing dementia patients' agitation. Caregivers were asked how confident they were about handling the problem for each identified behavioral problem and if they believed that they could manage the problem for behaviors that did not occur. Scores range from 42 to 210, with higher scores representing greater caregiver self-efficacy. In this study, Cronbach's alphas ranged from 0.98 to 0.99 at different time points.

Caregivers received general information on dementia care and follow-up phone calls simply to maintain contact, but without any training for developing a behavioral problem-management plan and strategies.