Evaluating the use of participatory action research to implement evidence-based guidance on dementia palliative care in long-term care settings: A creative hermeneutic analysis.

BACKGROUND: Dementia affects a large proportion of society and places a significant burden on older people and healthcare systems internationally. Managing symptoms at the end of life for people with dementia is complex. Participatory action research can offer an approach that helps to encourage implementation of evidence-based practices in long-term care settings.
METHODS: Three evidence-based guidance documents (pain assessment and management, medication management, nutrition and hydration management) were introduced in three long-term care settings for older people. Data generated from work-based learning groups were analysed using a critical hermeneutic approach to explore the use of participatory action research to support the implementation of guidance documents in these settings.
RESULTS: Engagement and Facilitation emerged as key factors which both enabled and hindered the PAR processes at each study site.
CONCLUSIONS: This study adds to the body of knowledge that emphasises the value of participatory action research in enabling practice change. It further identifies key practice development approaches that are necessary to enable a PAR approach to occur in care settings for older people with dementia. The study highlights the need to ensure that dedicated attention is paid to strategies that facilitate key transformations in clinical practice.

What does this research add to existing knowledge in gerontology?

This study enhances the body of knowledge on participatory action research approaches within long‐term care settings.

It outlines the use of work‐based learning as a strategy to facilitate learning in practice settings.

What are the implications of this new knowledge for nursing care with older people?

Facilitation is a valuable tool for supporting implementation of practice change in the care of older people.

Critically reflecting in and on practice enables a creative problem‐solving approach to nursing care challenges.

How could the findings be used to influence policy or practice or research or education?

Identification of practice context, competing agendas and ways to promote engagement is necessary when attempting to change practice and should be considered at the outset of any practice change initiative.

Implementation of guidance is necessary for practice enhancement and facilitation as a strategy using work‐based learning activities is effective, but further studies should evaluate the role of leadership in implementation projects.

INTRODUCTION

Dementia is a progressive illness, and the importance of palliative care has been increasingly acknowledged (Department of Health, 2020). By 2025, it is estimated that approximately 65,000 adults will be living with dementia in Ireland (Alzheimer Europe, 2020). Approximately 19,530 people with dementia resided in nursing homes in Ireland in 2016 (Pierse et al., 2019). As the condition progresses into the later stages, the person will have increased symptoms of pain, eating and swallowing, cognitive and behavioural symptoms, and infections (Sampson, 2010). Managing these symptoms raises ethical challenges as care shifts from life‐extending medical interventions to ‘comfort care’ approaches in advanced dementia (Sampson, 2010).

To address the complexity of the palliative care needs of people with dementia, there is an increasing focus on the importance of developing an evidence‐base and evidence‐based guidelines to support dementia palliative care. Against this background, the Irish Hospice Foundation and a number of collaborators developed evidence‐based guidance documents for the management of dementia palliative care. The documents targeted the management of symptoms and care challenges that commonly present in advanced dementia including pain assessment and management (Cornally et al., 2016); management of hydration and nutrition (Hartigan et al., 2016); and medication assessment and medication management (Lehane et al., 2016). In relation to each of the three areas, the documents provide guidance on appropriate assessment of the resident's current state and preferences, the involvement of the family when information cannot be obtained from the resident, and formulation of a management plan to support quality of care. The content of each guidance document is outlined in Table 1.

TABLE 1

Key components of the dementia palliative care guidance documents

Document	Focus of the guidance document
Pain assessment and management	• Recognising pain in dementia • Pain assessment in mild to severe dementia • Developing a pain management plan • Referral to specialist services
Management of hydration and nutrition	• Common Eating, Drinking and Swallowing Difficulties in People with Dementia • Assessment of Hydration and Nutrition • Management of Hydration and Nutrition • Decision‐Making – Balancing Different Views
Medication assessment and management	• Medication management of cognitive symptoms and maintenance of function • Medication management of non‐cognitive symptoms and behaviours that challenge • Optimising decision‐making through application of medication management principles • Medication administration

Extracted from dementia palliative care guidance documents (Cornally et al., 2016; Hartigan et al., 2016; Lehane et al., 2016).

However, the publication of guidelines or research findings does not guarantee adherence to the recommendations in practice (Matthew‐Maich et al., 2013; Snelgrove‐Clarke et al., 2015). Research has shown that imparting information is not sufficient to change practice (Chapman, 2016; Matthew‐Maich et al., 2013). Furthermore, education alone, which enhances nurse's knowledge and attitudes, has been found to have less impact than practice‐based initiatives on patient outcomes (Chapman, 2016; Gijbels et al., 2010; Matthew‐Maich et al., 2013). In the long‐term care setting (LTC) context, reviews indicate that interventions and guidelines have varied influence on staff behaviours and patient outcomes (Diehl et al., 2016; Low et al., 2015). A recent scoping review also sought to identify strategies to implement evidence‐based practice for palliative care in long‐term care settings, highlighting the challenge of implementation in this context (Collingridge Moore et al., 2020). Similarly, a systematic review found that no one single strategy, or combination of strategies, can be linked directly to successful implementation of nursing guidelines (Spoon et al., 2020). Implementation of evidence into practice is still a field in development and warrants further exploration (Kindblom et al., 2021).

A growing body of evidence suggests that action learning and participatory action research (PAR) may help to encourage evidence‐based practice beyond traditional methods of education. Action research offers a systematic and intentional approach to bring about change. The facilitation process, using a PAR methodology, creates a context where researchers and experienced facilitators interact at the level of the staff, supporting them to identify what they need to learn, set goals and modify patterns of care. PAR which encourages problem‐based reflections on behaviour and assumptions that interfere with individual learning and effective work performance (McNamara et al., 2014). A small number of studies have used PAR to implement palliative care for advanced dementia in long‐term care settings, taking various approaches to PAR (Andrews et al., 2009; Mitchell et al., 2020; Stacpoole et al., 2015). The PAR approach reported in this paper was part of a larger study that aimed to attend to both effectiveness and implementation of evidence in long‐term care settings for people with advanced dementia (Coffey et al., 2021; Timmons et al., 2021). The overarching project aim was to introduce guidance documents into practice, in three LTC sites, then to examine how the guidance influences palliative care for persons with dementia. Evidence suggests that dementia care education is associated with greater outcomes when components of PAR such as interactive group work, problem‐based learning, an experienced facilitator and the relevance of the intervention to practice are included (Surr et al., 2017). In this study, PAR in the form of work‐based learning groups (WBLGs) and facilitation was used to implement evidence‐based guidance on three areas of dementia palliative care. The aim of this paper is to report on the process of using PAR to implement guidance in long‐term care settings.

METHODS

Design

This study applied a participatory action research (PAR) approach (Damschroder et al., 2009; MacDonald et al., 2012). PAR was chosen as it maximises participation of participants and researchers in terms of co‐creating an understanding of facilitating the implementation of evidence‐based practices by way of determining issues, concerns, and conceptualising solutions. The process of implementing the ‘innovation’ (i.e guidance documents) through use of facilitation and work‐based learning is the focus of this paper. Key features of work‐based learning include the emphasis on experiential learning (Little & Brennan, 1996; Dewar & Walker, 1999) and facilitation of critical reflection in the creation of new professional knowledge (Clarke & Copeland, 2003, Gallagher & Holland, 2004). The sessions are also designed to meet the needs of the workplace and the learner (Swallow et al., 2001; Clarke & Copeland, 2003; Sobiechowska & Maisch, 2006).

Five WBLG sessions took place at each site (15 in total) over a 6‐month period, with session length ranging from 30 to 90 minutes approximately. The sessions were structured around how practice could be changed within the context of each setting to address the guidance documents. During the first WBLG session, the facilitators presented participants with a brief overview of relevant guidance documents and findings from the situational analysis (Timmons et al., 2021) prior to ‘ice‐breaker’ and ‘brain‐storming’ activities about priority areas. As the sessions progressed, facilitation activities such as ‘circle of concern/circle of influence’, ‘in and out’, helped in working through case study development and associated action plans on agreed practice change areas. An outline of these sessions (Discussion Topics and facilitation processes) can be seen in Table 2. All participants were encouraged and given time to express their views and perspectives, experiences, thoughts, insights, concerns and opinions with all voices given equal consideration. The expert action facilitator researcher provided prompt questions to ensure that the process did not go off track or beyond scope.

TABLE 2

Work‐based Learning Group Structure & Processes for Participatory Action Research Project‐ Implementing Evidence Based Guidance for Dementia Palliative Care

Session details	Structure/Outline of Reflective Workbased learning group	Processes used	Participants per session
Session 1. Getting to know guidance document and aspects of Dementia Care Agreeing ways of working. Introduction of guidance document	▪ Overview of guidance document and Dementia Care ▪ Development of guidance ▪ Linking guidance to dementia care ▪ How can this be achieved in practice? ▪ Critically looking at current workplace culture	▪ Presentation and facilitated discussion. ▪ “Claims, concerns and issues”. ▪ Creative session with participants asking them to address meta‐theme ▪ Agreeing an engagement contract.	Site 1 n = 3 Site 2 n = 4 Site 3 n = 4
Session 2. Identifying ways to promote guidance practice	▪ Re‐engagement with guidance recommendations and gaining a more in‐depth insight ▪ “Claims concerns and issues” ▪ Linking guidance to practice ▪ Outlining Action Plans	▪ Reflection on issues identified in “Claims, concerns and issues”. ▪ Creating a landscape of the workplace culture. ▪ Identifying strategies for improvement in relation to guidance evidence.	Site 1 = 7 Site 2 = 3 Site 3 = 4
Session 3 Using data collected to devise action plans.	▪ Work‐based learning activities‐ how did they go and what were the outcomes? ▪ Looking at strategies from Day 1 and discussion from Day 2 and devising action plans to be worked on over the next few sessions.	▪ Reflecting on learning and implications of identified practice. ▪ Development of an action plan. (what is an action plan and how do we use it?)	Site 1 = 4 Site 2 = 8 Site 3 = 3
Session 4. Making the guidance real	▪ Recap on guidance ▪ Recap on WBL activities. ▪ Discussion on data from practice observations and informal discussions carried out by staff with residents/families ▪ Action planning. What has been achieved? And how?	▪ Making the evidence real ‐ using it in everyday language and continuing to build knowledge of how the guidance works. ▪ Gain an understanding of how to use data collected. ▪ Looking at what you see happening/the way things are being done in the analysis and how things should be done ▪ Reflecting on learning and implications for ongoing activities, including the further development of action plans.	Site 1 = 5 Site 2 = 8 Site 3 = 4
Session 5. Evaluation	▪ Reflection on work based activities that have been taking place since last session ▪ Evaluation of taking part in the study	▪ Creative exercise to determine how everyone felt about taking part and also to look at changes in practice (if any)	Site 1 = 4 Site 2 = 2 Site 3 = 5

Ethical approval was granted from the University Clinical Research Ethics Committee (log number: ECM4 (oo) 5/6/18 & ECM 3(nnnn)3/7/18).

Setting and participants

The study took place in three long‐term care settings for older people in the Republic of Ireland. Each setting varied in terms of size and organisation with bed numbers ranging from 46 to 97. Study participants were interdisciplinary healthcare staff (Site 1 N = 43, Site 2 N = 24, Site 3 N = 26) including nurses, healthcare assistants (HCAs), catering staff, support staff, pharmacists, speech and language therapists, and dieticians. At each site, a project ‘champion’, who had a leadership or managerial role on each ward, for example Clinical Nurse Manager (CNM), recruited participants involved in the provision of care to residents. Attendance at the work‐based learning groups ranged from 2 to 8 participants across the 3 sites over 6 months. Each site had two external facilitators: the guidance topic lead and an expert action researcher. A site‐specific clinical practice development coordinator also participated.

Data generation

All facilitators and healthcare participants generated data derived from the work‐based learning groups including feedback from participants on work‐based learning activities, facilitator field notes and reflections directly after each WBLG. Specifically, data regarding healthcare provider experiences and context were captured by the WBLG session notes (n = 15) whereby staff were facilitated through creative and reflective exercises to express their views and consider their practices as related to the specific guidance area within their work environment, for example observations of practice and informal interviews with residents. Data in the form of meeting agendas (n = 15) and facilitator field notes (n = 15), which included a debriefing summary of the effectiveness of facilitation strategies used, were also collected. Researcher structured reflections were another data source (n = 15), to promote objectivity regarding knowledge construction (Polit & Beck, 2012). Researcher observations, thoughts/feelings and personal evaluation were documented regarding situational components and group dynamics. Final WBLG reflections were collated to capture additional insights into the action research process, particularly concerning those aspects of the process that presented challenges or enablers of practice change.

Data analysis

Data analysis from the WBLGs was concerned with both site‐specific evaluation and with overarching evaluation of the process. The analysis was both ongoing (happening at each WBLG session day with participants) and overarching (at the end of the study with the guidance leads and expert facilitator). The guidance leads, the expert facilitator and the project lead, analysed the data using a creative hermeneutic data analysis approach (Boomer & McCormack, 2010) see (Figure 1).

FIGURE 1

Creative hermeneutic data analysis (Boomer & McCormack, 2010)

The use of creative arts can lead to new interpretations and ways of working and in analysis can highlight patterns, themes and connections (Boomer & McCormack, 2010). Using this artistic method; the guidance leads, facilitator and project lead engaged in eight stages of individual and group analysis processes, as outlined in Figure 1, with the intention of developing an agreed set of themes among the whole group. The first steps involved all members of the group looking at the raw data and creating an image or creative expression of the data. They next told the story of their image to one other person in the group who wrote the story verbatim. The tellers and writers switched and repeated the process. The facilitators next themed their images. The group then came together and shared all the themes they had devised. The group categorised the themes and developed a set of group themes. Following these stages, agreed themes were developed, representing all the data and agreed by all group members. The final stage represented the group writing a ‘meta‐narrative’ representing all the themes. The researchers undertook this approach for each of the three sites individually to analyse site‐specific implementation and cumulatively to analyse the overall process.

The agreed themes and resulting meta‐narratives for each site are firstly presented, followed by findings of the overarching analysis. While the former highlighted the unique issues that both enabled and hindered implementation at individual sites, certain overlapping themes also emerged across sites. To capture this, an overarching evaluation was conducted.

Site 1 – Agreed themes and meta‐narrative

The analysis of data for site 1 following the creative hermeneutic process (Boomer & McCormack, 2010) as described in Figure 1 led to the development of 5 themes: Leadership, uncertainty, competing agendas, light bulb moments and meaningful engagement. These themes informed the metanarrative for site 1.

Metanarrative site 1

Initially people were uncertain about the process and outcome (both facilitators and staff). There were many competing agendas. Uncertainty reigned. All were fearful and had wavering attitudes and negativity was encountered at the start. Time was an issue because of competing agendas, resources and need for ongoing support. Flickers of progress interspersed throughout the spiral of change gave way to ‘light‐bulb’ moments, comprehension and problem solving. Enthusiasm, temporary collectiveness and the role of helping hands and ongoing support led to meaningful engagement. A thread of leadership was present but not visible at key junctures.

Leadership was identified as both an enabling and constraining factor with the researchers identifying how critical the role of the leader was to the implementation process. Staffing issues, increased responsibility and workload led to the unavoidable availability of the champion for the early sessions. This influenced staff engagement. While as many staff as possible were facilitated to attend the WBLGs, the non‐attendance of a key leader led to a “worry that the senior leadership was absent from the process” and the “lack of champion and continuity led to uncertainty”. Uncertainty was initially reflected through participants’ reticence about the project, with wavering attitudes expressed regarding their commitment. Staff could see the value of the guidance but were concerned that this would cause further work for them. Stress and worry at the beginning of the project led to negativity. It was ‘fairly flat… going nowhere at the beginning because ….people were wondering what moves they should make and how the whole thing hung together’. Resources, time and competing agendas were merged as themes to illustrate the pressures staff were experiencing in practice. Throughout the project, there was an element of ‘constraints versus hope’. Time proved a significant issue for both staff and facilitators. Staff found it hard to give time to the WBLG while on the ward due to busyness, with ‘pressure’ on facilitators to deliver the WBLGs within the time available. Lightbulb Moments reflected instances where ‘there were “aha!” moments of learning for staff engaged with the process’ and where people could see the benefit of not only using the document but also of using a problem‐solving and shared decision‐making approach ‐ ‘It was like the mist cleared and they were able to find a route for themselves’. Meaningful engagement occurred in the later WBLGs and represented a ‘temporary state of coming together’ of both staff and facilitators. As sessions progressed, staff could see the perceived value of the document and expressed enthusiasm both for the process and the guidance ‘which led to enlightenment and an upward swing in the use of the guidance’. It was particularly evident that the guidance of a skilled facilitator was necessary to provide a ‘helping hand’ but not ‘take charge’ of the implementation process. While staff were enthusiastic in completing action plans, by the end of the WBLGs, there was a recognition that ‘ongoing support’ may be necessary to ensure continued use of the guidance in practice.

Site 2 – Agreed themes and meta‐narrative

The themes of lack of awareness of purpose, leaning in looking on, breakthrough, fanning embers of interest and practice change constraints informed the metanarrative for site 2.

Metanarrative site 2

Lack of awareness of purpose was the “elephant in the room” that reduced in size over time but always remained. There was uncertainty about the road ahead but staff went with the flow although they lacked direction and continuous immersion. “Lone voyagers” versus “interested onlookers” represented differences between staff who were “leaning in” while others “looked on”. A period of percolation led to breakthroughs. Through slow and steady progress, reflecting and looking critically at practice led to “fanning the embers of interest” that harnessed a commitment to change. The turbulent nature of commitment secondary to competing demands led to practice change constraints.

The theme, lack of awareness of purpose, represents the initial uncertainty about the ‘road ahead’ and the purpose of the guidance. This was expressed by the ‘elephant in the room’ which was an undertone of reluctance by staff to commit to change. There was a lack of direction, awareness and continuity including ‘breaks in communication’ between staff who attended the sessions and those who did not ‐ ‘different people attended every session and did not appear to have understanding of WBLG activities …..…. felt a lot of confusion and inconsistencies with attendees’ As the sessions progressed, it was noted that staff either leaned in or looked on. There were the proactive few, the ‘lone voyagers’ who were using the guidance in practice while others were ‘interested onlookers’. While there was a supportive champion on site, their attendance at the WBLGs was inconsistent which led to a feeling of getting ‘stuck in the mud, ….2 steps forward 2 steps back and lack of ownership’. Nevertheless, a breakthrough in the form of ‘those one or two (staff) who were fully engaged and were a messenger to others about how guidance can support better practices’ helped to advance change, even though ‘it took a few sessions to embrace it’. Fanning the embers of interest reflects the outcomes of the processes taken within the WBLGs and the critical approach staff took in looking at their current practice. This enabled a ‘continuous flow of ideas, engagement and a way to neutralise any negativity’ and represented a way of creating energy ‘.…even though they hadn't understood the activities, there was huge energy’. Practice change constraints arose from the turbulent nature of commitment secondary to competing demands by both staff and facilitators. From the staff perspective ‘although it was seen as relevant it had turbulence at many junctures with regard to integration…. there was a reluctance to commit to change, staff were trying to open the door of change but were unable to because of being constrained with time and current documentation’. From a facilitator perspective, there was an initial ‘push/pull’ between being didactic and creative: ‘sometimes there was an emphasis or over ambition in terms of activities, which didn't allow for creativity to come through’.

Site 3 – Agreed themes and metanarrative

The five themes of security and willingness to change, deflection, caught in the Web, leadership Influence, and buds of growth informed the metanarrative for site 3.

Metanarrative site 3

Change is never easy. Key ingredients of security and willingness help staff understand the process and make individuals comfortable to share concerns and ideas. With facilitator effort, staff saw value in the guidance. A lack of understanding of the purpose led, at times, to deflection rather than reflection in practice. Being anchored to current practice led to a sense of defensiveness, vulnerability and fear of getting it wrong. Researchers were viewed as regulators rather than co‐practitioners. Preparatory work led to “buds of growth” with moments of enlightenment and improvement. Sustainability of these buds of growth was influenced by systems, struggles and leadership styles.

Security and Willingness to Change arose from reflection on evidence that staff were ‘trying to embrace change’ yet were continually ‘stuck’ and ‘chained’ to current practices. While staff saw value in guidance and were in tune with residents’ needs, there was a sense of vulnerability, which required ‘facilitator effort’ to address, before staff began to engage with the process and were comfortable to ‘express their knowledge, share their experiences or ideas’. From session to session, staff displayed both ‘intentional and unintentional ’ rather than ‘reflection’ and engagement with the process. Initially staff were defensive and used the WBLG sessions to ‘vent their discontent’. As the sessions progressed, facilitators reflected on a ‘spoon feeding’ rather than ‘facilitated support’ approach being required. The theme, caught in a web, represents the organisational constraints that both participants and facilitators felt during the implementation. There was a ‘sense of vulnerability and fear’ from staff with the facilitators being perceived as ‘imposters in practice’. There was a ‘glass half empty’ attitude portrayed in embracing the opportunity, and facilitators had to provide ‘lots of coaxing’ with ‘little return’. Leadership Influence conveyed through a ‘hierarchical management style’ and ‘lack of support’ from the champion meant there was a struggle to keep the project afloat. Staff displayed ‘subservient struggles’ with ‘time lags between sessions’ because of unexpected site problems. This impacted on continuity of learning activities and cast doubt on the sustainability of the implementation process overall. However, despite a thorny and prickly journey throughout the project there were buds of growth. The ‘activities and exercises were stimulating for all’ and made the guidance real for staff. Once barriers were lowered, there was ‘evidence of building blocks to engagement’ and ‘moments of enlightenment’ which showed that staff were in tune with patient needs and learning to understand the process which led to ‘good ideas for improvements’ being shared.

Overarching evaluation of the implementation at all three sites – Agreed themes and meta‐narrative

The overarching evaluation ensured that the process of implementation, and how that was facilitated were analysed with the intent of assessing their effectiveness in providing guidance going forward for future projects. This analysis was developed from the facilitators reflecting on the process and data gleaned from staff. Three themes, Journeying through, Patterns of Engagement and Constrained Willingness informed the overarching metanarrative.

Overarching metanarrative

During the journey, there were sparks of progress overshadowed by doubtfulness and constraints. There was a process of uncertainty in direction. The process would not have advanced without the external facilitators or without the champions. Constrained willingness was evident which resulted in insecurity and a lack of power. Staff were anchored to current practice. There were positive beginnings and when people were given the time to reflect and build on their reflection through incorporating new learning, they came out of themselves and blossomed. Without the push and pull there would not have been further progression.

Journeying through reflects the many factors involved in moving from point A to point B and progressing to key junctures in the implementation process which made the experience ‘invigorating, motivating or enlightening’. ‘Goal achievement’ resulted in a lot of excitement and delight and helped to ‘engender a sense of ownership’ to carry forward the implementation. Patterns of Engagement with the implementation process required ‘ structures’ and ‘support’. Where groups came together with cooperation and were set on the same course, this was seen as a ‘vessel of change’. This was underpinned by leadership style or ‘degrees of leadership’ and key factors such as ‘facilitators’. Leadership styles that were ‘rooted and tied to old ways of working’ often led to the leader being very visible or invisible. Both leadership style and expert external facilitators were key components to the success of the implementation. Constrained Willingness reflects the predominant initial uphill struggles and ‘daunting uncertainties’. ‘Elements of break‐through’, ‘problem solving’ and ‘troubleshooting’ led to ‘sparks of progression’. The many ‘moving parts’ such as ‘competing agendas, challenges in practice and lack of understanding or awareness’ led to differing approaches with ‘releasing control’ and ‘embracing change’. However, where this was embraced there was a sense of ‘integration’ with practice and a willingness to carry forward the change.

DISCUSSION

This study explored the process of using PAR to support the implementation of evidence‐based guidance in LTC settings for older people. Engagement, through participation by leaders and staff, and facilitation approaches, in the form of work‐based learning activities, were evidenced across all three sites and in the overarching evaluation.

Engagement is seen as important and necessary in action research (Snoeren e al., 2012). Engagement in action research is often discussed in terms of participation and involvement. In this study there were differing levels of engagement by staff champions and leaders. In practice, engagement was not always straightforward, and we encountered various levels of engagement throughout the project. Engagement was influenced by competing agendas of workload and time available for participation in the work‐based learning groups; this is similar to the findings of Snoeren et al. (2012). Fluctuating involvement of the champions and the revolving attendance of participants at work‐based learning days further hindered engagement. Grant (2004) advocates having a strong goal‐orientated leader to bring about a change in culture. In this study, clinical leaders on all sites were project champions. The extent of their involvement varied, some completed the majority of work themselves and did not facilitate staff participation, while others gave staff full reign but no direction. An integrative review of champions in healthcare‐related implementation by Miech et al. (2018) suggests that while champions alone are not enough to bring about change, having a champion was more likely to lead to success. Shaw et al. (2012) observed that champions need to be facilitative and must be able to empower staff and create environments where staff feel psychologically safe. Further, their study proposed that being a leader in clinical practice did not necessarily translate to effective practice change. This is similar to the findings of the present study, where staff were unsure of their roles, were constrained by current practices, and reported both being spoon‐fed and left to their own devices. It further underlines the need to research the role of the champion or leader in the role of practice change, and to define the leadership style of the champion prior to implementation as advocated by Buckley et al. (2018).

Work‐based learning groups provide the opportunity for critical reflection and communicative space for open, trustworthy and reciprocal relationships to be formed. Involvement led to opportunities for growth and breakthrough moments in the study. Helyer (2015) suggests that learning by reflection is crucial for practitioners to gain understanding and in turn insights that can improve practice. According to Williams (2010), work‐based learning can ensure that deep learning occurs, the result of which impacts on both the professional development of the learner and working practices. Reflections, both in and on practice and personal critical reflection, have long been advocated as a method of professional and personal development (Boud et al., 2013; Kolb, 1984; Rolfe, 1996; Schon, 1987). The WBLGs enabled participants to critically reflect on both the guidance documents and the process of implementation, which enabled action to occur, with guidance document implementation (to varying degrees) in all sites. Reflexive action research, where researchers and participants dialogically critique current ways of working and devise new practices collaboratively, provides opportunities for self and organisational development (Ripamonti et al., 2016). Using work‐based learning within a PAR approach, enabled the generation of knowledge between researchers and participants, allowed communicative spaces where issues could be discussed, and promoted the utilisation of guidance documents in practice. However, achieving balance in PAR can be difficult to accomplish (Jacobs, 2010). In our study, this was exemplified by the competing agendas of the researchers and participants that did not always align and affected engagement. While the researchers attempted to negate this by identifying assumptions and expectations at the outset of the project, we believe that early addressing of competing agendas is a critical stage of PAR and implementation research and recommend that further research be conducted in this critical area.

Expert facilitators were a key component to the success of the work‐based learning groups. They were seen as providing a ‘helping hand’ and facilitated support, which enabled participants to share their practice knowledge and effect change. The i‐PARiHS framework describes an expert facilitator as one who understands and is sensitive to contextual elements and can integrate context, innovation and participants (Harvey & Kitson, 2016). Both van der Zijpp et al. (2016) and Øye et al. (2016) argue that expert facilitation is necessary for successful implementation. This study's findings support these claims but further offers, as proposed by Hardiman and Dewing (2019), that a facilitator should not only have topic expertise but should also have facilitation expertise, that is the skills necessary to facilitate change. There is evidence from practice development initiatives in particular that expert facilitation is key to success (Hardiman & Dewing, 2019; Mekki et al., 2017; Raelin, 2012; Webster & Dewing, 2007). Indeed, Harvey et al. (2002) in their concept analysis of facilitation describe the purpose of facilitation as being to support, and to enable people to analyse current practice, which in turn leads to change in behaviour and work practices. This is borne out in our study where staff relied on the facilitators to enable them look critically at their current practice and to assist with identifying ways to implement the guidance document. Törmä et al. (2021) found that a facilitated implementation strategy based on action research approaches was necessary to bring about change in practice. We would agree with this finding and further hypothesise that an ‘expert’ facilitator and work‐based learning approaches further increase the probability of engagement of staff and identification of new ways of working.

Strengths and limitations

A number of steps were taken to establish data trustworthiness. Strategies to ensure credibility included democratic and sustained engagement with study participants throughout the project duration and the use of investigator and data triangulation. Specifically, a number of data sources (e.g. agendas, meeting minutes, facilitator debriefs and reflections) were triangulated. Investigator triangulation was also applied whereby several research team members were involved in addressing both the organisational aspects of the study as well as the process of analysis. To judge the extent to which the findings are transferable, it was important to adequately describe the study context, that is, participant characteristics and study setting. Reflexivity, in terms of examining one's own conceptual lens, explicit and implicit assumptions, preconceptions and values, and how these affect research decisions, was ongoing throughout the study. These took the form of debriefs, developing agendas in response to the previous WBLG needs, and reflections that research team members wrote after conducting the WBLGs. Participant values and assumptions were also ascertained in the WBLG days through facilitation exercises and crucially, it was the participants themselves who decided upon the problem areas to work with during the PAR sessions. Additional WBLGs over a more prolonged period would have been of benefit to assist in further embedding evidence‐based practices. This was reflected through healthcare staff requests for further support by the external facilitators after the final session.

CONCLUSIONS

This study has explored the process of using PAR to effect practice change in three core areas of dementia palliative care. In this instance, the approach taken was that of blending work‐based learning groups with the key practice development principles of facilitation to engage all relevant stakeholders in realising practical, ‘real‐world,’ changes at the point of care. To conclude, in adopting a PAR approach to guide change in dementia palliative care, designating dedicated attention to the process of implementation facilitates greater insights into what is required to enable ‘key transformations’ in clinical practice.

Implications for practice

An important implication for practice is the necessity of a skilled facilitator to optimise WBLG outputs especially in terms of fostering collaborative working and collective decision‐making.

A further practice implication relates to the importance of understanding the contextual needs of the setting and stakeholders. In being open to such investigation, barriers to practice change implementation can be circumnavigated.

Engagement by all stakeholders in the implementation process is important. Strategies such as work‐based learning, identification of leadership style and working with competing agendas need consideration when negotiating practice change.

CONFLICT OF INTEREST

The authors declare no conflict of interest.

AUTHOR CONTRIBUTIONS

All authors contributed to the study design. EL, NC, IH and CB conducted the work‐based learning groups. AC, EL NC IH and CB conducted the data analysis and the interpretation of data. CB and EL drafted the manuscript. All authors contributed, read and approved the final manuscript.