The three core objectives of the study and findings are set out below:

1. To understand the potential of using low-cost pet robots as a psychosocial intervention for older adults and people with dementia.
First, we reviewed the literature⁷ and identified 8 articles that evaluated the use of 2 low-cost pet robots with older adults and people with dementia. This included a robotic cat and dog, which costs between €110-130 (as of December 2020). These studies were conducted in the UK and USA. Some of the positive impacts of using pet robots included improvements to mood, communication, and social interaction. Some studies also reported that the use of pet robots provided companionship and led to improvements in mental wellbeing. However, issues and concerns associated with the use of pet robots were also reported. For instance, some older adults misperceived pet robots as live animals, or became attached to them. They also led to anxiety or agitation among a few participants, who attempted to harm the pet robot. Some care staff and researchers also had concerns about the hygiene and infection control of the pet robots.

Next, we reviewed 15 online consumer websites, and identified 1,327 reviews that described experiences and perceptions of using a low-cost robotic cat with older adults and people with dementia⁸. Reviewers (individuals who wrote the reviews on the consumer websites) were mostly family members of older adults and people with dementia. Many (n=357) described positive perceptions about the robotic cat’s lifelikeness and features, while some (n=105) felt that it felt and looked mechanical. Most (n=874) described that its use led to positive emotions among older adults and people with dementia, such as improved mood and expressions of affection. For instance, one reviewer shared “I would say this week has been his calmest, happiest, most relaxed, enjoyable week in possibly three or more years! Because of this life-like, mechanical companion designed exactly for people like him”. Reviewers also expressed that it provided the older adult and or person with dementia an opportunity to participate in various activities, such as grooming or talking to it. Nevertheless, some reviewers (n=70) expressed that it caused anxiety for those who saw it and treated it as being real.

Overall, these findings were similar with other larger studies^9-13 that used other higher-cost pet robots. This suggests that low-cost robotic pets show promise in addressing the psychosocial needs of older adults and people with dementia.

2. To understand barriers and facilitators to implementing pet robots for older adults and people with dementia.
We reviewed the literature to identify and consolidate information from 53 articles that reported on the barriers and facilitators to implementing social robots, including pet robots^{14, 15}. Most studies (n=37) were conducted within 13 countries in Europe. Other countries included Australia, the United Kingdom, the United States, New Zealand, Japan and Mexico. Most studies were conducted in participants’ homes and long-term care facilities. We found that previous research studies have been disproportionately focused on examining barriers and facilitators relating to the characteristics of social robots (including pet robots), such as their design and ease of use, while little is known about other contextual factors, such as external policies, that can influence their implementation. No previous studies have been conducted to explicitly examine the multi-level determinants to implementing social robots, including pet robots.

After this, we used a qualitative approach to further understand this topic by interviewing healthcare professionals and organisational leaders from 8 nursing homes in the Republic of Ireland^{16, 17}. Multi-level barriers and facilitators to the implementation of pet robots were identified.  Some barriers included cost, lack of funding, resources and knowledge, infection control mandates and potential disruptions to workflow. Some facilitators identified included relative advantages compared to other existing interventions, national guidelines, and alignment with care processes.

3. To identify, contextualise and gain consensus on strategies for implementing pet robots in nursing homes for dementia care.
Achieving this objective involved 2 main sequential steps¹⁸. Step 1 was to identify and contextualise implementation strategies. Findings of barriers and facilitators to implementing pet robots (Objective 2) were mapped onto a taxonomy of implementation strategies^{19, 20}, which allowed us to identify a list of potentially relevant strategies. These strategies were contextualised to the context of this study, using findings from our previous studies and through consultation with 2 healthcare professionals, 1 academic researcher and 1 organisational leader, and with a Patient and Public Involvement member from the Alzheimer Society of Ireland.

Step 2 was to gain consensus from a panel of experts. A two-round modified Delphi study will be conducted with 3 groups of key stakeholders to identify the strategies that are important and critical for implementing pet robots in long term care facilities as part of a care plan for residents with dementia. This process is ongoing and is expected to be completed by end of Summer 2022.