The aim of this study is to conduct and evaluate a controlled experimental trial aimed at exploring whether, to what extent and how a socially-assistive Pepper robot that operates on a unique CARESSES cultural competence solution can produce better health and well-being related outcomes among older adults residing in long stay care homes (and their informal carers) compared to a control socially-assistive Pepper robot with an alternative CARESSES solution, as well as care as usual.

The objective of the "Culture-Aware Robots and Environmental Sensor Systems for Elderly Support' (CARESSES) project is to build culturally competent robots that are able to re-configure their way of acting and speaking when offering a service to a person they are assisting, and match his / her culture, customs and etiquette. This study will test and evaluate robots that have been configured to provide a culturally-competent care to older adults residing in care homes. It is hypothesised that by introducing robots, which are more sensitive to the user's needs and able to adapt to his / her cultural background, the CARESSES' innovative solution will offer a safe, reliable and intuitive system that fosters independence and autonomy, improves quality of life of older adults, and also reduces informal carer's burden. The methodology adopted to develop this culturally competent solution is based on the principles of Transcultural Nursing, specifically on Hofstede's Cultural dimensions theory (for describing general cultural characteristics at national level) and on Papadopoulos, Tilki and Taylor's model of cultural competence.

At the heart of the CARESSES software is a cultural knowledge base (CKB) that stores all the required information about the user and his / her cultural background. CKB was developed using the guidelines provided by the experts in Transcultural Nursing, who identified through field observations differences and similarities between individuals within three cultural groups (i.e. White-English, Indian and Japanese) in order to avoid as much as possible stereotyping and to provide suggestions regarding how likely culturally-specific behaviours are valid for an individual belonging to the cultural group.

The study will employ a mixed method (qualitative and quantitative) single-blind, controlled before-and-after experimental design with two testing sites (UK and Japan), in which participants will either be purposefully allocated to an experimental arm (who will receive and be assisted by the CARESSES robot), control arm 1 (who will receive and be assisted by the control CARESSES robot) or control arm 2 (who will receive care as usual). Between and within subjects comparisons will be used to evaluate whether and how the assistance of a CARESSES Pepper robot improves health and well-being related outcomes among care home residents and their informal carers. Eligible older adults from three cultural groups (White-English, Indian and Japanese) will be recruited from the UK-based (White-English and Indian cultural backgrounds only) and Japanese-based testing sites. Additionally, the participants will be requested to nominate up to three informal carers of their choice who will also be invited to take part in the study, but to a different degree. After the recruitment, the participants from both sites will then be purposefully allocated to either an experimental arm (UK site: n=10; Japan site: n=5) that will adopt the CARESSES robot, or to a control arm 1 (UK site: n=10; Japan site: n=5) that will adopt the CARESSES control robot, or to control arm 2 (UK site: n=10; Japan site n=5).

In the UK site, each of the arms will include five participants who primarily identify themselves with the White-English culture, and five participants who primarily identify themselves with the Indian culture. In the Japan site, three arms will only include participants who primarily identify themselves with the Japanese culture. The participants allocated to the control arm 2 will engage in testing first, followed by the control arm 1 and then the experimental arm. Testing will occur on two participants at any one time in the UK site, and on one participant in the Japan site. Each participant will adopt a Pepper robot over a period of two weeks.

Prior to commencing the full-scale study, a pre-trial feasibility pilot will be conducted aimed at assessing the feasibility and acceptability of the screening, recruitment and evaluation procedures, as well as identifying and tackling any technical issues associated with the implementation of the CARESSES robot within a care home. One resident, who primarily identifies himself / herself with the White-English culture and one informal carer will be invited to take part in the pilot study. The evaluation procedure will take place in a condensed format across one week (9 hours) rather than two weeks (18 hours) as during the full-scale study.

Inclusion Criteria (Older Adults):

• Aged ≥ 65 years.
• Reside in an eligible Advinia care home in the UK or within the HISUISUI facility in Japan.
• Reside in a single occupancy bedroom / bedroom area.
• Identify themselves as primarily belonging to the White-English or Indian cultures (UK site only), or Japanese culture (Japan site only).
• Unlikely to express aggression towards themselves, the robot, and/or the researcher (as assessed by the interRAI- Long Term Care Facility (LTCF) tool Aggressive Behaviour Scale (ABS) < 1).
• Have the cognitive ability to participate in the study (as assessed by the interRAI-LTCF tool Cognitive Performance Scale (CPS) ≤ 2).
• Unlikely to fall ill and be hospitalized during study period (as assessed by the FRAIL-NH scale ≤ 10)
• Verbally able to communicate in and understand English (UK site only) or Japanese (Japan site only).

Exclusion Criteria:

• Aged < 65 years.
• Reside in multiple-occupancy bedrooms (UK site only)
• Do not identify themselves as primarily belonging to the Indian, White-English or Japanese culture.
• Likely to express aggression towards themselves the robot, and/or the researcher (as assessed by the interRAI-LTCF tool ABS ≥ 1).
• Do not have sufficient cognitive ability to participate in the study (as assessed by the interRAI-LTCF tool CPS > 2).
• Likely to fall ill and be hospitalized during study period (as assessed by the FRAIL-NH scale > 10)
• Not able to communicate in and understand English (UK site only) or Japanese (Japan site only).

Inclusion criteria (Informal carers)

• Aged ≥18 years.
• Have visited the participant in the care home within the past 3 months.
• Provide any type of informal help, care and/or support to the participant.
• Are a relative, partner, friend or neighbour who has a significant personal relationship with the participant.
• Are not paid or officially employed to provide care to the participant.
• Able to communicate in and understand English (UK site) or Japanese (Japanese site).

Exclusion criteria (Informal carers)

• Aged < 18 years.
• Have not visited the participant in the care home within the past 3 months.
• Not able to verbally communicate in and understand English (UK site only) or Japanese (Japan site only).
• Are formally paid to provide care to the participant.

The researchers and technical staff supporting the experiments are unable to be blinded. Therefore, a single blind masking strategy will be used meaning that the participants and their carers (apart from the control arm 2) will be left unaware of what group they have been allocated to and what type of the CARESSES robot they will be interacting with.