Family caregivers of persons with dementia (PwD) are well known to be "invisible patients." Existing studies have suggested that sense of coherence (SOC), a core inner strength, could be a protective factor against depression among family caregivers of PwD. However, few interventions are available to enhance the SOC and thereby the emotional and psychological health of this population. While in-home social robots are viewed as a useful addition to caring for PwD, the use of social robots with family caregivers as the primary audience are vastly under-investigated. In response, the investigators have developed a strength-based, robot-assisted intervention based on Antonovsky's theory of saluotogenesis for family caregivers of PwD. The feasibility, acceptability, and preliminary effects of the robot on family caregivers of PwD have been confirmed, with recommendations for examining its applicability among this population in a full trial.

This study aims to examine the effects of a salutogenic strength-based, robot-assisted intervention on improving depressive symptoms in family caregivers of PwD. This is a mixed-methods study, including a prospective, single-blinded, two-arm parallel randomized controlled trial and semi-structured individual interviews. A group of 148 family caregivers of PwD will be randomized to the intervention group (IG) or the control group (CG), who will receive reading materials on dementia care weekly. Each IG participant will engage with a robot integrated with an AI-powered chatbot at home for 6 weeks. Individual interviews will be conducted at week-6. Purposive sampling will be used to ensure diversity among the participants. Descriptive statistics, t-test and/or Chi square test, generalized estimating equation (GEE), linear regression and thematic analysis will be used.

Primary hypothesis: Participants in the intervention group (IG) will report a greater improvement in depressive symptoms than the control group (CG) immediately postintervention at 6 weeks (T1), 1-month (T2) and 3-month (T3) after completing the intervention.

Secondary hypothesis: 1) Participants in the IG will report a greater improvement in SOC, perceived caregiving burden, coping, and caregiving self-efficacy than the CG at T1, T2, and T3; 2) The frequency of engaging with specific robot features is associated with improved depressive symptoms, SOC, caregiving burden, coping, and caregiving self-efficacy.

The findings may: 1) advance the understanding of the application of the salutogenesis theory to robotic technology, 2) boost the inner strength of family caregivers to maintain their emotional and psychological health in stressful situations, and provide a new paradigm to strengthen the informal care system.

Inclusion Criteria:

• those aged 18 years or older;
• family caregivers of PwD who had been clinically diagnosed;
• are living in the same household as the PwD;
• providing daily caregiving for ≥ 4 hours /day;
• have mild-to-moderate depressive symptoms (PHQ-9 score of 5-14) (Yu et al., 2012); and
• are able to speak Cantonese.

Exclusion Criteria:

• those who have been diagnosed with a mental disorder such as bipolar disorder, schizophrenia, dementia, or depression in an acute stage;
• those with a visual or hearing impairment that would interfere with the use of a robot;
• those currently using a social robot at home; and
• who have limited access to electricity at home.