Namaste care in the home setting: developing initial realist explanatory theories and uncovering unintended outcomes

Introduction The End-of-Life Namaste Care Program for People with Dementia, challenges the misconception that people with dementia are a ‘shell’; it provides a holistic approach using the five senses, which can provide positive ways of communicating and emotional responses. It is proposed Namaste Care can improve communication and the relationships families and friends have with the person with dementia. Previously used in care homes, this study is the first to explore the pioneering use of Namaste Care in people’s own homes. Objective To develop initial programme theories detailing if, how and under which circumstances Namaste Care works when implemented at home. Design A qualitative realist approach following the RAMESES II guidelines was employed to understand not only whether Namaste Care has positive outcomes, but also how these are generated, for whom they happen and in which circumstances. Setting A hospice in the North East of England, operating in the community, through volunteers. Participants Programme theories were developed from three focus groups with volunteers implementing Namaste Care (n=8; n=8; n=11) and eight interviews with family carers (n=8). Results Four refined explanatory theories are presented: increasing engagement, respite for family carers, importance of matched volunteers and increasing social interaction. It was identified that while Namaste Care achieved some of the same goals in the home setting as it does in the care home setting, it could also function in a different way that promoted socialisation. Conclusions Namaste Care provides holistic and personalised care to people with both moderate and advanced dementia, improving engagement and reducing social isolation. In the present study carers often chose to use Namaste Care sessions as respite. This was often linked to their frustration of the unavoidable dominance of task-focussed care in daily life. Individualised Namaste Care activities thus led to positive outcomes for both those with dementia and their carers.

Two specialist workers lead the project and orchestrate training, debrief events and matching of patients and volunteers. Volunteers visit the person for twenty sessions, which are usually weekly and last two hours. Delivery is therefore significantly different to that initially outlined by the originator, who suggests that it should be delivered twice a day, seven days a week  ( Table 1). However, stakeholders in a recent review indicated that this was unlikely to be feasible in most care homes in the UK . The review also found little empirical evidence on the optimal 'dose' of sensory interventions, such as NC, although the literature did suggest that interventions that are delivered more regularly are important for creating a sense of reassurance and familiarity and building trusting relationships between residents and carers.
Home delivery of the intervention also differs significantly from care home delivery in terms of staff impact; use of NC in care homes is also intended to address staff satisfaction by enabling them to have quality time with residents that is not just focused on task-based activities. However, there are similar implications for family members' in the delivery of NC in the home environment, as volunteers delivering NC encourage their participation. This would engage family members in quality time with their loved one, as opposed to task focused care.    This research contributes in two ways to the NC nascent knowledge base. Whilst research to date has demonstrated outcomes in care homes, little is yet understood about how and why they occur.
In addition, the unique implementation in a community setting affords the opportunity to explore the impact of home as a novel intervention context.

Methods
Realist evaluation is a theory driven approach which seeks to understand not only whether an intervention works, but what it is about it that works, for whom, in what circumstances and why (Pawson & Manzano-Santaella, 2012). It acknowledges that interventions take place within complex social systems  and is therefore well suited to studying interventions such as NC.
The formulae Context + Mechanism = Outcome (C+M=O) is used to express this. An Cunningham, 2018), the use of a realist approach will help to expose the multiple resources delivered as part of NC, the ways that these may be employed with different people, in diverse situations, and how these generate outcomes. Applying the principles of realist evaluation therefore will determine why NC is successful or unsuccessful, in particular contexts.
This research was approved through Northumbria University Ethical Approval System (reference: HLSCW161705). All participants gave informed consent. Due to the exploratory and small scale nature of the research patients and the public were not involved.
A realist approach was operationalized in two phases following the RAMESE II guidelines surrounding the development of programme theory: phase 1 focused on building programme theories with volunteers implementing NC in the community, using focus groups (n = 3, with 8, 8 and 11 participants respectively); phase 2 consisted of refining the theories with family carers of people who had received the NC intervention (n=8). Focus groups took place at the hospice and interviews were conducted either at the hospice (n=1) or at the family home (n=7). All focus groups and interviews were digitally recorded. Participants were recruited through the Namaste Leads. Volunteers and family carers had the study explained to them in person by the Namaste Lead; if they were interested in participating they provided their email address and/or telephone number with permission for it to be given to the lead researcher (SD). SD then contacted potential participants to arrange a suitable time and location for interview (family carers) or provided the date and time of the focus group (volunteers). Participation in the focus groups and interviews was voluntary and attendance at one focus group did not assume attendance at subsequent ones.

Setting and referrals
The hospice is set in the North East of England and covers two areas, one town (population of around 25,000) and one city (population of around 65,000). The hospice delivering Namaste in the community was founded in 1988 and is a registered charity which also receives some income from the National Health Service (NHS). The mission of the hospice is to make every day count for those with life limiting illnesses. Their vision is to be a centre of excellence within the community and to provide compassionate and individualised care in the right time and place for the person.
Family carers self-referred to the hospice to request access to NC. They were then matched with a trained volunteer. The hospice received requests for NC from family carers of people with severe and milder dementia. In order to be inclusive, as a community intervention, the hospice provided NC to all, not just to those with advanced Dementia. Referral criteria is provided as supplementary information 1. Volunteers visited the person with dementia twenty times, in their own home, once per week for 2 hours. Should volunteers encounter issues they reported immediately to one of the two NC leads, one of whom was a trained Admiral Nurse. NC sessions were personalised based on the person's 'Life Story', which was completed before NC sessions began by the NC Lead and shared with the volunteer. All sessions included multisensory bespoke activities such as hand massage, aromatherapy and music in those with more advanced dementia, and exploring the garden, baking and singing in those with milder dementia.
Data was transcribed verbatim and imported into NVivo. A realist logic of analysis employed CMOC was used to build and refine programme theory (  Volunteers are referred to throughout analysis as V1-V12, and family carers as P1 -P8.
Objective: To develop initial programme theories detailing if, how and under which circumstances NC works when implemented at home.

Findings
The findings are presented following the phases of the research, with outcomes stated clearly at the end of each phase.
Phase 1: Building programme theories

Impact on People with Dementia (PwD)
Volunteers were introduced to their matched PwD and their carer through the hospice NC Lead.
During this informal meeting, the PwD's Life Story was discussed, in the form of a larger document called 'My Namaste Care'. This formed a starting point for creating personalised care based on sensory interactions. This was a key step in matching personalities, histories and interests, which was thought to be significant to the intervention success. From the findings presented above, the following programme theory was developed: Programme theory 1: The volunteer is aware of the person's life story (context).
Experimentation based on the life story is used to identify useful personalised activities (resource) which evoke an emotional response from the person with dementia, meaning they engage with the NC worker (reasoning). The outcome can be relaxation, engagement, increase in alertness or emotional response.

Impact on family carers
One of the guiding principles of NC is to engage the people surrounding the PwD, whether this be care home staff or family carers. Volunteers suggested that often family carers felt that they had no hope and felt a sense of helplessness, which was compounded by a lack of support. Whilst professionals emphasised the need to trigger an emotional connection with the PwD, regardless of their verbal abilities, family carers talked more about the value of social interaction.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   23 One family carer in particular questioned whether it was specifically interaction with the NC volunteer that was important, or whether it was just social interaction in general.

Discussion
This preliminary study developed initial programme theories for the novel use of NC in peoples' own homes, as opposed to care homes. Including contrasting programme theories 2a and 2b, in total five programme theories were developed from the focus groups with NC volunteers, of these programme theories, 3 were supported (table 2). As dementia progresses, people's opportunities to engage in social interactions that are meaningful to them become more limited (context). Using their knowledge of the person's life story to develop a set of bespoke interactional tools and techniques (resources), NC volunteers evoke an emotional response in the PwD (reasoning), leading to a set of relaxation, engagement and alertness outcomes. 2a.
In a context where carers have seen their loved one decline and been told there is 'no hope' and received little or impersonal care (context) use of NC to evoke reactions from their loved one (resource) leads to them feeling hopeful and acknowledging that their loved one is still 'living' (reasoning). This leads to increased hope (outcome) and wellbeing.
Carers provide task-focused care and have little input from other services (context). A familiar NC volunteer provides 2 hours of interaction with the PwD (resource) which eases off worries about the PwD and allows them to have some respite (reasoning) which leads to an increase in wellbeing (outcome).
Carers provide continuous care and have little input from other services (context), provision of 2 hours contact with a trained NC volunteer (resource) allows them to concentrate on other things, knowing that the PwD is in safe hands (reasoning) which gives them restorative time and space (outcome).

3.
Being often task focused, family members recognise the value of NC (context). A tailored activity box is left at the person's home (resource). The family are keen to engage in activities that enable them to connect emotionally with the PwD (reasoning  Image reproduced with permission of the rights holder, Professor Mansfield-Cohen. This exploratory research has started to provide explanations of how NC may work in the home setting. Future research has been briefly previously outlined, but could also include investigations of use of volunteers to deliver NC in care homes, to allow the intervention to also be delivered to those with milder dementia. Furthermore, an ethnographic approach to develop further understanding of outcomes for those receiving NC would be beneficial.

Strengths and Limitations
To our knowledge, this is the first formal evaluation of NC in the home setting. It is also the first to explore the use of volunteers to deliver NC. The findings highlight that further research is necessary, but outline interesting findings in terms of intervention fidelity and unintended outcomes.
A limitation of the study is the sample size; while some programme theories were not substantiated by the data, it could be that this was due to this particular sample. As in all realist research, these findings do not claim finality, but merely the beginning of an explanatory endeavour for NC. As is the process for realist research, theories were tested and refined or rejected. We aimed to report as much as possible on the process of analysis in order to be transparent and rigorous.
Furthermore, it is important to counteract publication bias of only positive results, although we do not consider the unintended consequences identified in this study negative (that of respite).
Finally, it also enables the research field to build on the knowledge created and discourages repeated research in the same area.

Implications for clinicians and policymakers
The research highlights positive outcomes for PwD, volunteers and family members. However, it also highlights that NC may not work in the same way in the persons own home, as it does care homes. This does not detract from the value of NC, but warrents further investigation. It also indicates the unmet needs of family carers. In order to facilitate PwD to live at home and to meet the current drive of care into the community, we need to firstly ensure the needs of those with dementia and their carers are met, whether these needs be physical, emotional or social.   Image reproduced with permission of the rights holder, Professor Mansfield-Cohen. • The person living with dementia is most likely in their last year of life.

Conclusion
• The person living with dementia is finding it more difficult to communicate verbally.
• They have become completely dependent on the support of others for activities of daily living.
• They would not now find it easy to leave the house or engage in group activities.
• They would benefit from a gentle, sensory approach, on a one to one basis by a trained volunteer to enhance their wellbeing.
• The person with dementia and/or carer has consented to the referral and is aware that the carer needs to be present in the house during Namaste visits. The views expressed in this paper do not necessarily represent those of the funders or UKCRC.
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing Interests
The authors declare no competing interests.

Data Statement
Data from the study is not shared as participants consented for their data to be used only in this study.

Abstract
The End-Of-Life Namaste Care Program for People with Dementia, challenges the misconception that people with dementia are a 'shell'; it provides a holistic approach using the five senses, which can provide positive ways of communicating and emotional responses.
Previously used in care homes, this study is the first to explore the pioneering use of Namaste Care in people's own homes.
Objective: To develop initial programme theories detailing if, how and under which circumstances Namaste Care works when implemented at home.

Design:
A qualitative realist approach following the RAMESES II guidelines was employed to understand not only whether Namaste Care has positive outcomes, but also how these are generated, for whom they happen, and in which circumstances.
Setting: A hospice in the North East of England, operating in the community, through volunteers.
Participants: Programme theories were developed from three focus groups with volunteers implementing Namaste Care and eight interviews with family carers.

Strengths and limitations of this study
 This small and specialist study explores the use of a novel intervention, Namaste Care, in a new setting, the person's own home.
 The article reports two stages related to theory building using a realist approach in order to be rigorous and transparent in the theory development surrounding Namaste Care use in the home setting.
 A limitation of the study is the sample size; while some programme theories were not substantiated by the data, it could be that this was due to the limited sample size.
 Vast differences are acknowledged in the implementation of Namaste Care in people's own homes in comparison to care homes. disease progresses often the focus of care shifts toward the physical body. [6][7][8][9] This emphasis on physical needs often comes at the expense of personhood needs. [10] Activity has been demonstrated to be a positive therapeutic intervention with potential to enhance quality of life and reduce behavioural symptoms in those with dementia, thus potentially avoiding pharmacological treatments. [11] There is an increasing body of research into nonpharmacological, psychosocial and community-based interventions and their impact on quality of life and well-being for people with dementia and their family members or carers. [12][13][14] Accordingly matched with a person with dementia, in terms of personality, abilities and interests, for example.
Two specialist workers lead the project and orchestrate training, debrief events and matching of patients and volunteers. Volunteers visit the person for twenty sessions, which are usually weekly and last two hours. Delivery is therefore significantly different to that initially outlined by the originator, who suggests that it should be delivered twice a day, seven days a week [26] (Table   1). However, stakeholders in a recent review indicated that this was unlikely to be feasible in most care homes in the UK. [19] The review also found little empirical evidence on the optimal 'dose' of sensory interventions, such as NC, although the literature did suggest that interventions that are delivered more regularly are important for creating a sense of reassurance and familiarity and building trusting relationships between residents and carers. Home delivery of the intervention also differs significantly from care home delivery in terms of staff impact; use of NC in care homes is also intended to address staff satisfaction by enabling them to have quality time with residents that is not just focused on task-based activities. However, there are similar implications for family members' in the delivery of NC in the home environment, as volunteers delivering NC encourage their participation. This would engage family members in quality time with their loved one, as opposed to task focused care.
To our knowledge, this is one of only two hospices in the UK implementing this type of model for NC; the other service is located in London. Whilst this study is focused on delivery of NC in the person's own home, it will highlight pivotal contexts (not just related to physical location) and underlying mechanisms, which may also be relevant to the care home setting. The context and mechanisms identified in this research could warrant further research in the care home setting. Secondly, the unique implementation in a community setting affords the opportunity to explore the impact of home as a novel intervention context.
Objective: To develop initial programme theories detailing if, how and under which circumstances NC works when implemented at home.

Methods
Realist evaluation is a theory driven approach which seeks to understand not only whether an CMO configurations are used as explanatory formulae (otherwise referred to as realist programme theories), which are developed and refined with empirical data. As with other evaluations of person centred interventions, [33] the use of a realist approach will help to expose the multiple resources delivered as part of NC, the ways that these may be employed with different people, in diverse situations, and how these generate outcomes. Applying the principles of realist evaluation therefore will determine why NC is successful or unsuccessful, in particular contexts.

Ethical approval
This research was approved through Northumbria University Ethical Approval System (reference: HLSCW161705). All participants gave informed consent.

Patient and Public Involvement (PPI)
Due to the small-scale nature and limited funding of the research, patients and the public were not involved in the development of the research question or design of the study. Members of the public form the hospice were consulted on dissemination plans.

Operationalisation of the study
A realist approach was operationalized in two phases following the RAMESE II guidelines surrounding the development of programme theory: phase 1 focused on building programme theories with volunteers implementing NC in the community, using focus groups (n = 3, with 8, 8 and 11 participants respectively, 1 male in each focus group); phase 2 consisted of refining the theories with family carers of people who had received the NC intervention (n=8, 6 male, 2 female). Focus groups took place at the hospice and interviews were conducted either at the hospice (n=1) or at the family home (n=7). All focus groups and interviews were digitally recorded. Participants were recruited through the Namaste Leads. Volunteers and family carers had the study explained to them in person by the Namaste Lead; if they were interested in participating they provided their email address and/or telephone number with permission for it to be given to the lead researcher (SMD). SMD then contacted potential participants to arrange a suitable time and location for interview (family carers) or provided the date and time of the focus group (volunteers). Participation in the focus groups and interviews was voluntary and attendance at one focus group did not assume attendance at subsequent ones. Participants were not given any remuneration for the participation in the study, although volunteers were provided with lunch at the focus groups.

Setting and referrals
The hospice is set in the North East of England and covers two areas, one town (population of around 25,000) and one city (population of around 65,000). The hospice delivering Namaste in the community was founded in 1988 and is a registered charity, which also receives some income from the National Health Service (NHS).
Family carers self-referred to the hospice to request access to NC. The iterative approach adopted in realist evaluation allows the revisiting of the data as new additional questions emerge and connections are established, thus deepening the understanding and meaning of the findings. [36] Volunteers are referred to throughout analysis as V1-V12, and family carers as P1 -P8. The source of the data is indicated using Focus Group (FG) and then the number of the focus group (out of 3). For example, the first focus group is referred to as 'FG1'.

Findings
The findings are presented following the phases of the research, with outcomes stated clearly at the end of each phase. From the findings presented above, the following programme theory was developed: Programme theory 1: The volunteer is aware of the person's life story (context).
Experimentation based on the life story is used to identify useful personalised activities (resource) which evoke an emotional response from the person with dementia, meaning they engage with the NC worker (reasoning). The outcome can be relaxation, engagement, increase in alertness or emotional response.

Impact on family carers
One of the guiding principles of NC is to engage the people surrounding the person with dementia, As a result of this analysis, an alternative hypothesis was created: Programme theory 2b: Family carers provide task focused care and have little input from other services (context). A familiar NC volunteer provides 2 hours of interaction with the person with dementia (resource) which eases off worries about the family carer's loved one and allows them to have some respite (reasoning) which leads to an increase in well-being (outcome).

Family Carer use of NC
As described above, volunteers described how they felt often family carers roles had become task focussed, as opposed to engaging in enjoyable activities with their loved one. This was despite volunteers offering participation to family members. Family carers had shown initial interest in It was also evident that volunteers built up a very strong emotional connection with the person they were matched with. The following programme theory was built based on the analysis presented above:  between the volunteer and the person with dementia). The one on one delivery of NC in the home setting was also highlighted by family carers.

Discussion
This preliminary study developed initial programme theories for the novel use of NC in peoples' own homes, as opposed to care homes. Including contrasting programme theories 2a and 2b, in total five programme theories were developed from the focus groups with NC volunteers, of these programme theories, 3 were supported (table 2). The volunteer is aware of the person's life story (context). Experimentation based on the life story is used to identify useful personalised activities (resource) which evoke an emotional response from the person with dementia, meaning they engage with the NC worker (reasoning). The outcome can be relaxation, engagement, increase in alertness or emotional response.
As dementia progresses, people's opportunities to engage in social interactions that are meaningful to them become more limited (context). Using their knowledge of the person's life story to develop a set of bespoke interactional tools and techniques (resources), NC volunteers evoke an emotional response in the person (reasoning), leading to a set of relaxation, engagement and alertness outcomes. 2a.
In a context where carers have seen their loved one decline and been told there is 'no hope' and received little or impersonal care (context) use of NC to evoke reactions from their loved one (resource) leads to them feeling hopeful and acknowledging that their loved one is still 'living' (reasoning). This leads to increased hope (outcome) and wellbeing.
Not supported. Carers provide task-focused care and have little input from other services (context). A familiar NC volunteer provides 2 hours of interaction with the person with dementia (resource) which eases off worries about the person with dementia and allows them to have some respite (reasoning) which leads to an increase in well-being (outcome).
Carers provide continuous care and have little input from other services (context), provision of 2 hours contact with a trained NC volunteer (resource) allows them to concentrate on other things, knowing that the their loved one is in safe hands (reasoning) which gives them restorative time and space (outcome).

3.
Being often task focused, family members recognise the value of NC (context). A tailored activity box is left at the person's home (resource). The family are keen to engage in activities that enable them to connect emotionally with the person (reasoning).
Family use NC independently (outcome) Not supported 4.
One volunteer is aligned to a person with dementia and spends 2 hours per week solely with that person (context). The volunteer therefore has a knowledge history of what works/doesn't work and what the person likes (resources). This allows the volunteer and the person with dementia to develop a strong emotional connection (reasoning). The outcome could be considered as the recognition of the volunteer by the person with dementia but actually this leads to an increased engagement which might have previously been thought of as impossible (outcome).
One volunteer is aligned to a person with dementia and spends 2 hours per week solely with that person (context). The volunteer therefore has a knowledge history of what works/what doesn't work and what the person likes. (resources). This allows the volunteer and the person with dementia to develop a strong emotional connection (reasoning). The outcome could be considered as the recognition of the volunteer by the person with dementia but actually this leads to friendship, which could suggest an increased quality of life for both people (outcome).
The 'one on one' delivery of NC in the home setting in this study was highlighted by family carers as being preferable, not only because the person was in familiar surroundings but due to the increased engagement this provided. Family carers suggested that their loved one would be more likely to disengage in a group environment. NC aims to engage the senses and using it in the home setting could have the potential to allow more tailored delivery, with fewer distractions. Evidence suggests that sustained lack of stimulation can be detrimental to people in care homes who suffer from dementia, as it augments the apathy, boredom, depression, and loneliness that often accompany the progression of dementia. [37,38] The same, if not more enhanced, could be assumed for those with dementia who live at home and this could be supported by the preliminary findings of this research. This study and others [39,40] have highlighted the importance of social interaction for people living with dementia; those living at home with dementia have very little interaction with people other than their family and formal carers, due to issues of mobility and anxiety outside of home. Furthermore, family carers expressed an inability to interact with their loved one as they used to, this is in line with observations from another study using NC, which focused on touch. [20] This finding could warrant further investigation in care homes also.
Cohen-Mansfield et al. [38] suggest a framework for engagement of people with dementia ( Figure   1, reproduced). The theoretical framework suggests that environmental attributes (home setting), stimuli attributes (sensory activities) and person attributes (NC: Life story, matched volunteers and continuity with volunteer), alongside interactions among these attributes, affect engagement with stimuli by the person who has dementia. NC in the home environment could be said to be more open to personalised and tailored activities than a care home environment, with a 'one on one' approach and less distractions, such as other residents, therefore making the environment facilitative. Stimuli presented to people with dementia in NC are also matched at first with the person's attributes, through use of the Life Story. Cohen-Mansfield et al. [19] suggest that personalised activities are more likely to engage those with dementia. [41] This conceptual framework concerning engagement of persons with dementia therefore reflects NC well and could also be applicable to the use of NC in care homes, as well as in people's own homes. The authors have also developed a measurement of engagement, which could potentially be used in future research on NC given their complementarity of one another.  Image reproduced with permission of the rights holder, Professor Mansfield-Cohen.
Caring for people with dementia can be stressful, lead to family conflicts and cause burnout; [42] recent research has highlighted a need for further exploration of family carers' views about care for those with dementia at home. [43,44] One of the unintended consequences of NC in the home setting was its use by family carers for respite. NC aims to engage the family, with care home staff encouraging family and friends to join in where appropriate. [23,45] However, usual use of NC is in a care home setting, where family members do not provide the majority of task focused care. The family carers in this study described a lack of support and a need for respite, which is This exploratory research has started to provide explanations of how NC may work in the home setting. Future research has been briefly previously outlined, but could also include investigations of use of volunteers to deliver NC in care homes, to allow the intervention to also be delivered to those with milder dementia. Furthermore, an ethnographic approach to develop further understanding of outcomes for those receiving NC would be beneficial.

Strengths and Limitations
To our knowledge, this is the first formal evaluation of NC in the home setting. It is also the first to explore the use of volunteers to deliver NC. The findings highlight that further research is necessary, but outline interesting findings in terms of intervention fidelity and unintended outcomes.
A limitation of the study is the sample size; while some programme theories were not substantiated by the data, it could be that this was due to this particular sample. As in all realist research, these findings do not claim finality, but merely the beginning of an explanatory endeavour for NC.
A caution should also be outlined in interpreting the findings, due to the vast differences in implementation in people's own homes in comparison to care homes. Adapting an intervention like NC to work in the home environment does bring challenges for evaluation as the intervention itself is inevitably altered to facilitate delivery. In this delivery of the intervention, the 'dose' was different, however, recent research found little empirical evidence on the optimal 'dose' of sensory interventions. Furthermore, the interaction with volunteers as opposed to care home staff warrants further investigation and the inclusion of those with mild dementia poses questions around intervention focus and benefit, given that NC was developed for people with dementia who have physical and cognitive deterioration and are unable to engage with other activities.
However, recent research highlights the challenge of examining whether the impact of interventions vary depending on cognitive ability and indicates that further research is needed to assess how psychosocial interventions can be of use across the stages of dementia. [14] As is the process for realist research, theories were tested and refined or rejected. We aimed to report as much as possible on the process of analysis in order to be transparent and rigorous.
Furthermore, it is important to counteract publication bias of only positive results, although we do not consider the unintended consequences identified in this study negative (that of respite).
Finally, it also enables the research field to build on the knowledge created and discourages repeated research in the same area.

Implications for clinicians and policymakers
The research highlights positive outcomes for people with dementia, volunteers and family members. However, it also highlights that NC may not work in the same way in the persons own home, as it does care homes. This does not detract from the value of NC, but warrants further investigation. It also indicates the unmet needs of family carers. In order to facilitate those with dementia to live at home and to meet the current drive of care into the community, we need to firstly ensure the needs of those with dementia and their carers are met, whether these needs be physical, emotional or social.

Conclusion
A recent cohort study indicated that people with advanced dementia still often live with distressing symptoms [47] and that community services are often not tailored to their nonmedical needs. [29] Longitudinal input focused on improving quality of life using personalised interventions such as NC shows promise in optimising life for those with dementia and also could provide much needed respite for family carers when delivered in the home setting using volunteers.  Image reproduced with permission of the rights holder, Professor Mansfield-Cohen. • The person living with dementia is most likely in their last year of life.

Author Contributions
• The person living with dementia is finding it more difficult to communicate verbally.
• They have become completely dependent on the support of others for activities of daily living.
• They would not now find it easy to leave the house or engage in group activities.
• They would benefit from a gentle, sensory approach, on a one to one basis by a trained volunteer to enhance their wellbeing.
• The person with dementia and/or carer has consented to the referral and is aware that the carer needs to be present in the house during Namaste visits.

Data Statement
Data from the study is not shared as participants consented for their data to be used only in this study.  Design: A qualitative realist approach following the RAMESES II guidelines was employed to understand not only whether Namaste Care has positive outcomes, but also how these are generated, for whom they happen, and in which circumstances.
Setting: A hospice in the North East of England, operating in the community, through volunteers.
Participants: Programme theories were developed from three focus groups with volunteers implementing Namaste Care (n=8; n=8; n=11) and eight interviews with family carers (n=8).
Results: Four refined explanatory theories are presented: increasing engagement, respite for family carers, importance of matched volunteers and increasing social interaction. It was identified that whilst Namaste Care achieved some of the same goals in the home setting as it does in the care home setting, it could also function in a different way that promoted socialisation.
Conclusions: Namaste Care provides holistic and personalised care to people with both moderate and advanced dementia, improving engagement and reducing social isolation. In the present study carers often chose to use Namaste Care sessions as respite. This was often linked to their frustration of the unavoidable dominance of task-focused care in daily life. Individualised  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   5 Namaste Care activities thus led to positive outcomes for both those with dementia and their carers.

Keywords: Dementia, Namaste Care, Volunteers, Social Interaction, Respite
Article Summary

Strengths and limitations of this study
 This study details the theory building process in realist evaluation.
 Theory building was focused upon as opposed to theory testing, due to the lack of current evidence surrounding the use of Namaste in the person's own home and the small participant numbers.
 The study uses focus groups and interviews to develop rigorous and transparent programme theories  A limitation of the study is the sample size; while some programme theories were not substantiated by the data, it could be that this was due to the limited sample size.

Introduction
Globally, the numbers of people living with dementia will increase from 50 million in 2018 to 152 million in 2050, a 204% increase. [ 1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   6 Activity has been demonstrated to be a positive therapeutic intervention with potential to enhance quality of life and reduce behavioural symptoms in those with dementia, thus potentially avoiding pharmacological treatments. [11] There is an increasing body of research into nonpharmacological, psychosocial and community-based interventions and their impact on quality of life and well-being for people with dementia and their family members or carers. [12][13][14] Accordingly dementia. [18] NC is a psychosocial intervention that has been implemented variably internationally; [18] research is beginning to develop understanding about the intervention and it's cost implications, [19][20][21][22][23][24][25] but to our knowledge has only been formally evaluated in care home settings. A hospice in the North East of England has made provisions to provide NC in the person's own home. This is operationalised through the training of volunteers who are then matched with a person with dementia, in terms of personality, abilities and interests, for example.
Two specialist workers lead the project and orchestrate training, debrief events and matching of  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   7 patients and volunteers. Volunteers visit the person for twenty sessions, which are usually weekly and last two hours. Delivery is therefore significantly different to that initially outlined by the NC originator, who suggests that it should be delivered twice a day, seven days a week [26] ( Table 1). However, stakeholders in a recent review indicated that this was unlikely to be feasible in most care homes in the UK. [19] The review also found little empirical evidence on the optimal 'dose' of sensory interventions, such as NC, although the literature did suggest that interventions that are delivered more regularly are important for creating a sense of reassurance and familiarity and building trusting relationships between residents and carers. Home delivery of the intervention also differs significantly from care home delivery in terms of staff impact; use of NC in care homes is also intended to address staff satisfaction by enabling them to have quality time with residents that is not just focused on task-based activities. However, there are similar implications for family members' in the delivery of NC in the home environment, as volunteers delivering NC encourage their participation. This would engage family members in quality time with their loved one, as opposed to task focused care.
To our knowledge, this is one of only two hospices in the UK implementing this type of model for NC; the other service is located in London.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  Healthcare provision in Europe, the USA and Australia has seen an emphasis on providing people with choice around the location of their care and death, frequently with an emphasis on driving care into the community and facilitating home deaths. [27] Despite this, statistics indicate that home deaths in people with dementia are generally low internationally, with significant variance across countries reported as a product of variability in end of life care provision. [28] Furthermore, unmet needs are common in those with dementia living in the community, and most are nonmedical. [29] Recent research has highlighted that home-based dementia care should identify and address unmet needs by focusing on both care recipients and caregivers to enable the person with dementia to remain at home. [29] With current policy driving care into the community, ways to support quality of life for people with dementia in their own homes is pivotal.
This research contributes in two ways to the NC nascent knowledge base. Whilst research to date has demonstrated outcomes in care homes, little is yet understood about how and why they occur.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   9 Whilst this study is focused on delivery of NC in the person's own home, it will highlight pivotal contexts (not just related to physical location) and underlying mechanisms, which may also be relevant to the care home setting. The context and mechanisms identified in this research could warrant further research in the care home setting. Secondly, the unique implementation in a community setting affords the opportunity to explore the impact of the home as a novel intervention context.

Objective:
To develop initial programme theories detailing if, how and under which circumstances NC works when implemented at home.

Methods
Realist evaluation is a theory driven approach which seeks to understand not only whether an intervention works, but what it is about it that works, for whom, in what circumstances and why. [30] It acknowledges that interventions take place within complex social systems [31] and is therefore well suited to studying interventions such as NC.
The formulae Context + Mechanism = Outcome (C+M=O) is used to express this, with mechanisms consisting of both intervention resources and stakeholder reasoning. [32] An intervention offers resources (Mechanism resource: such as hand massage, for example) which can alter the context into which it is introduced [32] (C; the person with dementia is experiencing restlessness and agitation), triggering a change in the reasoning of intervention participants (Mechanism reasoning; patient relaxes and feels more able to engage), leading to a particular outcome (O; the person with dementia is less agitated potentially avoiding a respite admission).
CMO configurations are used as explanatory formulae (otherwise referred to as realist programme theories), which are developed and refined with empirical data. As with other evaluations of person centred interventions, [33] the use of a realist approach will help to expose  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   10 the multiple resources delivered as part of NC, the ways that these may be employed with different people, in diverse situations, and how these generate outcomes. Applying the principles of realist evaluation therefore will determine why NC is successful or unsuccessful, in particular contexts.

Ethical approval
This research was approved through Northumbria University Ethical Approval System (reference: HLSCW161705). All participants gave informed consent.

Patient and Public Involvement (PPI)
Due to the small-scale nature and limited funding of the research, patients and the public were not involved in the development of the research question or design of the study. Members of the public from the hospice were consulted on dissemination plans.

Operationalisation of the study
A realist approach was operationalized in two phases following the RAMESES II guidelines surrounding the development of programme theory: phase 1 focused on building programme theories with volunteers implementing NC in the community, using focus groups (n = 3, with 8, 8 and 11 participants respectively, 1 male in each focus group); phase 2 consisted of refining the theories with family carers of people who had received the NC intervention (n=8, 6 male, 2 female). Focus groups took place at the hospice and interviews were conducted either at the hospice (n=1) or at the family home (n=7). All focus groups and interviews were digitally recorded. Participants were recruited through the Namaste Leads. Volunteers and family carers had the study explained to them in person by the Namaste Lead; if they were interested in participating they provided their email address and/or telephone number with permission for it to be given to the lead researcher (SMD). SMD then contacted potential participants to arrange a suitable time and location for interview (family carers) or provided the date and time of the  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   11 focus group (volunteers). Participation in the focus groups and interviews was voluntary and attendance at one focus group did not assume attendance at subsequent ones. Participants were not given any remuneration for the participation in the study, although volunteers were provided with lunch at the focus groups.

Setting and referrals
The hospice is set in the North East of England and covers two areas, one town (population of around 25,000) and one city (population of around 65,000). The hospice delivering Namaste in the community was founded in 1988 and is a registered charity, which also receives some income from the National Health Service (NHS).
Family carers self-referred to the hospice to request access to NC. They were then matched with a trained volunteer. The hospice received requests for NC from family carers of people with severe and milder dementia. In order to be inclusive, as a community intervention, the hospice provided NC to all, not just to those with advanced Dementia. Referral criteria is provided as supplementary information 1. All family carers currently engaged with NC at the hospice at the time of the study were invited to participate, by telephone call conducted by one of the NC Leads.
Before interviews with family carers could be conducted, their loved one must have experienced a minimum of 4 NC sessions. This requirement, combined with the hospice's referral criteria constituted the inclusion and exclusion criteria for the study.
Volunteers all began training in NC three months prior to the study beginning. They were introduced to their matched person with dementia and their carer through the hospice NC Lead at the person's home. During this informal meeting, the Life Story of the person with dementia was discussed, in the form of a larger document called 'My Namaste Care'. This formed a starting point for creating personalised care based on sensory interactions. This was a key step in matching personalities, histories and interests, which was thought to be significant to the The iterative approach adopted in realist evaluation allows the revisiting of the data as new additional questions emerge and connections are established, thus deepening the understanding and meaning of the findings. [36] Volunteers are referred to throughout analysis as V1-V12, and family carers as P1 -P8. The source of the data is indicated using Focus Group (FG) and then the number of the focus group (out of 3). For example, the first focus group is referred to as 'FG1'.

Findings
The findings are presented following the phases of the research. From the findings presented above, the following programme theory was developed: Programme theory 1: The volunteer is aware of the person's life story (context).
Experimentation based on the life story is used to identify useful personalised activities (resource) which evoke an emotional response from the person with dementia, meaning they engage with the NC worker (reasoning). The outcome can be relaxation, engagement, increase in alertness or emotional response.

Impact on family carers
One of the guiding principles of NC is to engage the people surrounding the person with dementia, whether this be care home staff or family carers. Volunteers suggested that often family carers felt that they had no hope and felt a sense of helplessness, which was compounded by a lack of support. As a result of this analysis, a rival programme theory was created: Programme theory 2b: Family carers provide task focused care and have little input from other services (context). A familiar NC volunteer provides 2 hours of interaction with the person with dementia (resource) which eases off worries about the family carer's loved one and allows them to have some respite (reasoning) which leads to an increase in well-being (outcome).

Discussion
This preliminary study developed initial programme theories for the novel use of NC in peoples' own homes, as opposed to care homes. Including contrasting programme theories 2a and 2b, in total five programme theories were developed from the focus groups with NC volunteers, of these programme theories, 3 were supported (table 2).

The volunteer is aware of the person's life story (context). Experimentation based on the life story is used to identify useful personalised activities (resource)
which evoke an emotional response from the person with dementia, meaning they engage with the NC worker (reasoning). The outcome can be relaxation, engagement, increase in alertness or emotional response.

As dementia progresses, people's opportunities to engage in social interactions that are meaningful to them become more limited (context). Using their knowledge of the person's life story to develop a set of bespoke interactional tools and techniques (resources)
, NC volunteers evoke an emotional response in the person (reasoning), leading to a set of relaxation, engagement and alertness outcomes. 2a.
In a context where carers have seen their loved one decline and been told there is 'no hope' and received little or impersonal care (context) use of NC to evoke reactions from their loved one (resource) leads to them feeling hopeful and acknowledging that their loved one is still 'living' (reasoning). This leads to increased hope (outcome) and wellbeing.
Carers provide task-focused care and have little input from other services (context). A familiar NC volunteer provides 2 hours of interaction with the person with dementia (resource) which eases off worries about the person with dementia and allows them to have some respite (reasoning) which leads to an increase in well-being (outcome).
Family use NC independently (outcome) Not supported 4.
One volunteer is aligned to a person with dementia and spends 2 hours per week solely with that person (context). The volunteer therefore has a knowledge history of what works/doesn't work and what the person likes (resources). This allows the volunteer and the person with dementia to develop a strong emotional connection (reasoning). The outcome could be considered as the recognition of the volunteer by the person with dementia but actually this leads to an increased engagement which might have previously been thought of as impossible (outcome).
One volunteer is aligned to a person with dementia and spends 2 hours per week solely with that person (context). The volunteer therefore has a knowledge history of what works/what doesn't work and what the person likes. (resources). This allows the volunteer and the person with dementia to develop a strong emotional connection (reasoning). The outcome could be considered as the recognition of the volunteer by the person with dementia but actually this leads to friendship, which could suggest an increased quality of life for both people (outcome).
The 'one on one' delivery of NC in the home setting in this study was highlighted by family carers as being preferable, not only because the person was in familiar surroundings but due to the increased engagement this provided. Family carers suggested that their loved one would be more likely to disengage in a group environment. NC aims to engage the senses and using it in the home setting could have the potential to allow more tailored delivery, with fewer distractions.
Evidence suggests that sustained lack of stimulation can be detrimental to people in care homes who suffer from dementia, as it augments the apathy, boredom, depression, and loneliness that often accompany the progression of dementia. [37,38] The same, if not more enhanced, could be assumed for those with dementia who live at home and this could be supported by the preliminary findings of this research. This study and others [39,40] have highlighted the importance of social interaction for people living with dementia; those living at home with dementia have very little  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   32 interaction with people other than their family and formal carers, due to issues of mobility and anxiety outside of home. Furthermore, family carers expressed an inability to interact with their loved one as they used to, this is in line with observations from another study using NC, which focused on touch. [20] This finding could warrant further investigation in care homes also.
Cohen-Mansfield et al. [38] suggest a framework for engagement of people with dementia ( Figure   1, reproduced). The theoretical framework suggests that environmental attributes (home setting), stimuli attributes (sensory activities) and person attributes (NC: Life story, matched volunteers and continuity with volunteer), alongside interactions among these attributes, affect engagement with stimuli by the person who has dementia. NC in the home environment could be said to be more open to personalised and tailored activities than a care home environment, with a 'one on one' approach and less distractions, such as other residents, therefore making the environment facilitative. Stimuli presented to people with dementia in NC are also matched at first with the person's attributes, through use of the Life Story. Cohen-Mansfield et al. [19] suggest that personalised activities are more likely to engage those with dementia. [41] This conceptual framework concerning engagement of persons with dementia therefore reflects NC well and could also be applicable to the use of NC in care homes, as well as in people's own homes. The authors have also developed a measurement of engagement, which could potentially be used in future research on NC given their complementarity of one another. Image reproduced with permission of the rights holder, Professor Mansfield-Cohen.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   33 Caring for people with dementia can be stressful, lead to family conflicts and cause burnout; [42] recent research has highlighted a need for further exploration of family carers' views about care for those with dementia at home. [43,44] One of the unintended consequences of NC in the home setting was its use by family carers for respite. NC aims to engage the family, with care home staff encouraging family and friends to join in where appropriate. [23,45] However, usual use of NC is in a care home setting, where family members do not provide the majority of task focused care. The family carers in this study described a lack of support and a need for respite, which is This exploratory research has started to provide explanations of how NC may work in the home setting. Future research has been briefly previously outlined, but could also include investigations into volunteer delivery of NC in care homes, to allow the intervention to also be delivered to those with milder dementia. Furthermore, an ethnographic approach to develop further understanding of outcomes for those receiving NC would be beneficial.

Strengths and Limitations
To our knowledge, this is the first formal evaluation of NC in the home setting. It is also the first to explore the use of volunteers to deliver NC. The findings highlight that further research is necessary, but outline interesting findings in terms of intervention fidelity and unintended outcomes.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  A limitation of the study is the sample size; while some programme theories were not substantiated by the data, it could be that this was due to this particular sample. As in all realist research, these findings do not claim finality, but merely the beginning of an explanatory endeavour for NC.
A caution should also be outlined in interpreting the findings, due to the vast differences in implementation in people's own homes in comparison to care homes. Adapting an intervention like NC to work in the home environment does bring challenges for evaluation as the intervention itself is inevitably altered to facilitate delivery. In this delivery of the intervention, the 'dose' was different, however, recent research found little empirical evidence on the optimal 'dose' of sensory interventions. Furthermore, the interaction with volunteers as opposed to care home staff warrants further investigation and the inclusion of those with mild dementia poses questions around intervention focus and benefit, given that NC was developed for people with dementia who have physical and cognitive deterioration and are unable to engage with other activities.
However, recent research highlights the challenge of examining whether the impact of interventions vary depending on cognitive ability and indicates that further research is needed to assess how psychosocial interventions can be of use across the stages of dementia. [14] As is the process for realist research, theories were tested and refined or rejected. We aimed to report as much as possible on the process of analysis in order to be transparent and rigorous.
Furthermore, it is important to counteract publication bias of only positive results, although we do not consider the unintended consequences identified in this study negative (that of respite).

Implications for clinicians and policymakers
The research highlights positive outcomes for people with dementia, volunteers and family members. However, it also highlights that NC may not work in the same way in the persons own home, as it does in care homes. This does not detract from the value of NC, but warrants further investigation. It also indicates the unmet needs of family carers. In order to facilitate those with dementia to live at home and to meet the current drive of care into the community, we need to firstly ensure the needs of those with dementia and their carers are met, whether these needs be physical, emotional or social.

Conclusion
A recent cohort study indicated that people with advanced dementia still often live with distressing symptoms [47] and that community services are often not tailored to their nonmedical needs. [29] Longitudinal input focused on improving quality of life using personalised interventions such as NC shows promise in optimising life for those with dementia and also could provide much needed respite for family carers when delivered in the home setting using volunteers.
• The person living with dementia lives at home in the central [location details] or [location details].
• The person living with dementia is most likely in their last year of life.
• The person living with dementia is finding it more difficult to communicate verbally.
• They have become completely dependent on the support of others for activities of daily living.
• They would not now find it easy to leave the house or engage in group activities.
• They would benefit from a gentle, sensory approach, on a one to one basis by a trained volunteer to enhance their wellbeing.