Communication partner training for student health and social care professionals engaging with people with stroke acquired communication difficulties: A protocol for a realist review.

Stroke acquired communication impairments

In 2019 the global incidence of stroke was reported to be 12.2 million and stroke was reported as the third leading cause of death and disability resulting in 143 million people living with “disability-adjusted life-years" (Stark et al., 2021, p. 795). Among these life adjusting disabilities are the communication impairments aphasia, dysarthria, apraxia of speech and cognitive communication disorders (Baker et al., 2022). These impairments (overviewed in Table 1) can occur in isolation or in varying combinations and with varying severities. Their incidence and prevalence will increase in the coming decades in line with the predicted increased incidence of stroke world-wide (Stark et al., 2021). Specific to the European Union, Wafa et al. (2020) anticipate that the people living with stroke will have increased by 27% between 2017 and 2047. Studies reporting on the specific incidence and prevalence of stroke acquired communication impairments are variable, for example, they differ in geographical origins, number of participants and designs and methods (Frederick et al., 2022; Mitchell et al., 2021). Hence, there are variations across reports with aphasia impacting 7% to 77%, dysarthria impacting 24% to 69% and a combination of dysarthria and aphasia impacting 4% to 29% (Ali et al., 2015; De Cock et al., 2020; Dickey et al., 2010; Flowers et al., 2013; Frederick et al., 2022; Mitchell et al., 2021; Stipancic et al., 2019). Data on incidence of stroke acquired apraxia of speech is sparse and challenging to report on given that acquired apraxia of speech rarely exists in isolations but typically co-occurs with aphasia and/or dysarthria (Duffy, 2020). Regarding cognitive communication disorders associated with acute stroke, Riepe et al. (2004) reported that up to 77% exhibited cognitive impairment.

The stroke acquired communication impairments outlined in Table 1 present chronic, multifaceted challenges for the individuals presenting with the impairment/s and for those interacting with them (Ali et al., 2015; Chang et al., 2018; Mitchell et al., 2021; Wray et al., 2019). They can mask the individual’s inherent competency and result in communicative interactions that are negative for all interlocutors (Carragher et al., 2020; Kagan et al., 2001; Kagan et al., 2018; van Rijssen et al., 2021). Unsuccessful communication predisposes patients to increased risk of adverse hospital events (Bartlett et al., 2008; Hemsley et al., 2013); loss of autonomy and exclusion from participation in care related decisions (Brady et al., 2011; Carragher et al., 2020; O’Halloran et al., 2012). Additionally, there is increased risk of depression and anxiety (Shehata et al., 2015; Zanella et al., 2023).

When communication with patients is challenging, health and social care professionals (i.e., professionals with formal clinical training, e.g., doctors, nurses, physiotherapists, psychologists) can experience emotions such as frustration, impatience and guilt (Carragher et al., 2020; Hur & Kang, 2022). Consequently, they can limit their time interacting with people with communication impairments (Carragher et al., 2020). Student health and social care professionals (SH&SCPs) who may also, in contexts where professional registration is required, be referred to as pre-registration professionals, experience similar emotions. Over the course of their clinical training these students may be exposed to a variety of pedagogies and engage in a range of clinical activities and placements involving people with stroke acquired communication difficulties. Rathiram et al. (2022, p. 5) reported that SH&SCPs felt “emotionally strained, frustrated and helpless” when they could not understand their patients. In order to prevent these negative experiences health and social care professionals and SH&SCPs want to learn how to communicate successfully with people with communication difficulties (Carragher et al., 2020; Hur & Kang, 2022). Furthermore, people with aphasia believe that training health care professionals in the use of communication strategies is important in enabling them to live successfully with the communication impairment (Manning et al., 2019). CPT can provide the necessary learning opportunities.

Communication Partner Training (CPT)

CPT is defined by Cruice et al. (2018, p. 1) as an “umbrella term for a complex, behavioural intervention” that has many interacting components that are delivered in flexible ways. It is also described as an environmental intervention (Simmons-Mackie et al., 2010; Simmons-Mackie et al., 2016) as people around the person with communication impairment/s (i.e., the communication partners) adjust their behaviour and use communicative strategies and resources to facilitate information exchange (Simmons-Mackie et al., 2010; Simmons-Mackie et al., 2016). Such strategies and resources may include non-verbal cues such as gesture and pointing; personally relevant communication books and folders; applications on smart devices; writing; visual aids (e.g., maps, photo diaries) and modified verbal expression (e.g., slowed rate, using key words and short phrases). CPT can be successfully employed with multiple communication partners including health and social care professionals, SH&SCPs, family members, volunteers and people with communication impairments (Beeke et al., 2018; Cameron et al., 2018; Forsgren et al., 2017; Isaksen et al., 2023; Kagan et al., 2001; Rayner & Marshall, 2003). Importantly, CPT has the potential to pre-empt communicative challenges for our future frontline staff and enhance future clinical care for patients with stroke acquired communication impairments.

Employing realist review in CPT research

Realist terminology relevant to this realist review and all other phases of this research project is presented in Table 2 below.

Realist review is a systematic, “theory driven, interpretative” approach (Duddy & Wong, 2023, p. 1; Jagosh et al., 2014, p. 131) that facilitates the synthesis and evaluation of data of diverse methodologies and origins. It focuses on combining empirical evidence with theoretical understanding (Schick-Makaroff et al., 2016) in programme theory development. As Wong et al. (2012, p. 93) state, the premise underpinning realist review is that a specific intervention or class of interventions “trigger particular mechanisms somewhat differently in different contexts”, hence, as is the case in CPT research, outcomes vary across contexts. During programme theory development context, mechanism and outcome configurations (CMOCs) are employed to coherently explain how and why contextual variables influence intervention outcomes. Ontologically rooted in realism and aligned with Bhaskar’s stratified reality (Bhaskar, 1997), realist review explores both the visible and hidden forces that generate the outcomes of interest (Jagosh et al., 2014). From a realist standpoint, a behavioural intervention such as CPT is conceptualised to operate in open systems where the intervention changes the system and the system changes the intervention (Pawson et al., 2004; Pawson et al., 2005). Realist review aims to address the real world complexity and fluidity of these operating systems by going beyond simply asking if an intervention works and exploring generative causation to determine for whom it works, under what conditions, to what extent, how and why? Employing a realist approach has the potential to enhance existing knowledge and practice in stroke specific CPT for SH&SCPs. It can build on the current evidence base, which is predominantly impairment (aphasia) and context (chronic status) specific (Simmons-Mackie et al., 2016). It can address persisting theoretical and implementation gaps by explaining the relationship between specific contextual variables and the mechanisms they trigger to enact the reasoning and responses that bring about the intended or unintended outcomes (Dalkin et al., 2015; Wong et al., 2013).

Research on stroke specific CPT predominantly focuses on Aphasia (Chang et al., 2018) with an established evidence base underpinning recommendations for use of CPT for people with chronic aphasia (Simmons-Mackie et al., 2010; Simmons-Mackie et al., 2016). This includes established evidence for use of CPT with medical students (Legg et al., 2005). However, given that there is a high incidence of varied and co-occurring post stroke impairments across the acute to chronic care continuum, these recommendations, while very essential in the management of aphasia, do not comprehensively address the prevailing clinical realities of health and social care professionals or SH&SCPs working with people post stroke. Chang et al. (2018), surveying CPT practices of 122 Australian speech pathologists found that clinicians implementing CPT in clinical stroke settings adapt aphasia specific CPT programmes for use with the broader array of impairments. Correspondingly, in the literature there is increasing recognition of the need to develop CPT for a broader range of acquired impairments that address clinical reality (Chang et al., 2018; O’Rourke et al., 2018; Simmons-Mackie et al., 2016; Tessier et al., 2020). Studies addressing CPT for the broader range of stroke acquired communication impairments with SH&SCPs are emerging (e.g., Baylor et al., 2019; Burns et al., 2017; Forsgren et al., 2017; Mach et al., 2022). While the number of studies is small, they represent an important evolution in CPT development for SH&SCPs and this review aims to build on and potentially enhance this development. In realist review, demi-regularities or “semi-predictable patterns of occurrence” (Cunningham et al., 2021, p. 4) are explored across interventions and domains (such as health care and education) to uncover “families of mechanisms” (Pawson, 2002, p. 344) in programme theory development. Adhering to realist philosophy, it is these “families of mechanisms” (e.g., incentivisation, persuasion) rather than “families of interventions” that trigger change and enact outcomes (Pawson, 2002, p. 344; Wong et al., 2012, p. 94). By building on current knowledge and uncovering relevant “families of mechanisms”, this review, may enhance and potentially accelerate development and implementation of stroke specific CPT for SH&SCPs in clinical education.

The need to explore an approach to data synthesis that can accommodate all valuable research data is highlighted by Simmons-Mackie et al. (2016). In their updated systematic review, the authors reported on the emergence of studies addressing the efficacy of CPT use with the broader range of stroke acquired communication impairments and a promising new trend of studies, attempting to manage clinical reality and the feasibility of CPT implementation in complex settings. However, all of these studies were rated too low on the American Academy of Neurology levels of evidence (2011) for efficacy or effectiveness to provide recommendations for clinical practice. Given that realist review accommodates the synthesis and evaluation of data of diverse methodologies and origins, it can build on existing empirical findings from systematic reviews; overcome the challenge of accommodating all valuable research findings in the final analysis, and provide clinicians and educators with augmented theoretical and practical guidance for the development and implementation of CPT.

The value of theory driven CPT interventions and implementation strategies is increasingly recognised in the literature (Chang et al., 2018; Cruice et al., 2018; Shrubsole et al., 2023). However, theoretical approaches are not in mainstream use and given the vast array of behaviour change and implementation theories in existence it is a challenge for researchers to select the best fit (Eccles et al., 2012; Shrubsole et al., 2019). Shrubsole et al. (2019) argue that the potential contribution of different theories is unclear given that they have not been consistently applied to health professional’s behaviour or in this case, we argue, SH&SCPs’ behaviour. Also, as McGowan et al. (2020) point out, selecting only one or a few theories in behaviour change research and intervention puts the researcher at risk of omitting relevant factors. Notably, frameworks such as the Behaviour Change Wheel (BCW) (Michie et al., 2011) and the Theoretical Domains Framework (TDF) (Atkins et al., 2017; Cane et al., 2012; Michie et al., 2005), developed from synthesising theories and constructs, are being used more frequently in CPT research and other general research in acquired communication disorders (Behn et al., 2020; Chang et al., 2018; Johnson et al., 2017; van Rijssen et al., 2021). A realist approach considers existing substantive theories in programme theory formation. Consequently, it has the potential to overcome “best fit” challenges while building on existing theoretical understanding. Also, unlike other theoretical approaches, which can be limited in their ability to address multiple pertinent questions, realist review applies realist logic to answer the necessary range of pertinent questions - what works, for whom, under what conditions, to what extent, how and why? (Pawson et al., 2004; Wong et al., 2012; Wong et al., 2013). Answering such questions can strengthen current CPT research and practice by explaining “the success, failure” and “mixed fortunes” of this complex intervention (Wong et al., 2013, p. 1). Importantly, addressing these questions allows us to look beyond determining efficacy to determining necessary solutions for real world practice in higher education institutions. The anticipated outcome of this review, a middle range programme theory (defined in Table 2) ,while abstract, will be close enough to observable data to be usable and enable empirical testing.

Patient and Public Involvement (PPI)

PPI refers to research that is conducted ‘with’ or ‘by’ members of the public rather than ‘to’, ‘about’ or ‘for’ them (https://www.invo.org.uk/). It involves “working in collaboration or partnership with patients, carers, service users or the public” (https://hseresearch.ie/patient-and-public-involvement-in-research/). They may become “members of the research team, or part of reference groups, involved in key discussions and decisions, sharing their unique knowledge, expertise and perspective” (Staniszewska et al., 2018, p. 1). Originating from the disability and advocacy movements, it emphasises knowledge sharing between “lay people” and “the professionals” (McMenamin et al., 2022, p. 31). PPI benefits the research process, the researcher and the PPI contributors and there is increasing expectation for PPI in health and social care research (Hersh et al., 2021; McMenamin et al., 2022). It improves quality, relevance, impact and integrity of the research process. It promotes waste avoidance and embraces active involvement including shared decision making across the research cycle (Dawson et al., 2020; Hersh et al., 2021; McMenamin et al., 2022; Staniszewska & Denegri, 2013). It promotes inclusion, validation and understanding of the lived experience (Pearl & Cruice, 2017; Turk et al., 2017).

Incorporating PPI in realist review

While engagement of experts and stakeholders is commonplace in a realist review (Saul et al., 2013), the inclusion of PPI is at an evolutionary stage (Abrams et al., 2021). There is variability in all terms used to describe the realist review’s advisors or contributors and PPI tends to be subsumed into pre-existing structures or groups including “stakeholder, expert or advisory panels” (Abrams et al., 2021, p. 243). In this review, to support clarification for the reader, we use the term PPI advisor to denote people with stroke acquired communication impairments and their significant others. They, in collaboration with content advisors, relaist advisors, student advisors and an educationalist comprise this realist review’s advisory panel (see methods, Table 3).

Level of PPI advisors’ involvement in this realist review

In general, PPI can occur at different levels (e.g. consultation, defining scope, collaboration) and at different stages (e.g. design, project advisor, dissemination) of a research cycle (McMenamin et al., 2022). Specific to realist review Abrams et al. (2021) found that PPI contributor roles included informing and developing programme theory and advising over the course of the review. Developing research questions, sourcing literature and data extraction tend to be performed by advisors such as academics or clinicians. These traditional roles are reflected in the level of involvement expected from this review’s advisory panellists (see methods, Table 3). More broadly, across different types of research, there are a range of conceptual models overviewing different levels of involvement and power sharing in PPI, many of which have been influenced by Arnstein’s Ladder of Public Participation (1969) (McMenamin et al., 2022). In this realist review, we adopt the PPI Spectrum of Involvement in Research Model (IHRF, 2015 https://hrci.ie/about-us/our-work/ppi/). This model, depicted in Figure 1, illustrates that the PPI advisors on this review are involved at a consultative level.

Figure 1. PPI Spectrum of Involvement in Research Model (IHRF, 2015 https://hrci.ie/about-us/our-work/ppi/).

This figure has been reproduced with permission.

Aims and objectives

This review protocol overviews a realist review, the first phase of a PhD project. Figure 2 overviews the realist review in the context of the wider project. The wider project aims to employ a realist approach, including realist review (Phase 1) and realist evaluation (Phase 3), in combination with PPI (Phase 1- PPI advisors on review’s advisory panel and Phase 2- PPI Advisory Group). The primary aim of this realist review (Phase 1) is the synthesis of secondary data from varied relevant sources into a plausible and coherent middle range theory. This theory will explain how CPT is expected to work when teaching SH&SCPs to communicate with people with stroke acquired communication impairments. As stated above, the middle range theory developed will be close enough to observable data to be usable and permit empirical testing. It will undergo further appraisal and development during realist evaluation, the third phase of this project, in consultation with the PPI advisors who are consulting across all phases of this project.

A protocol for this review has been registered on PROSPERO on 01/05/2023 (CRD42023418951).

Figure 2. Project flow chart.

Steps in this realist review

While there are guiding principles around conducting a realist review there is no one prescribed method (Hunter et al., 2022) and this can result in variability across reviews and lack of clarity on key stages in the review process (Booth et al., 2019). This protocol aims to present a clear and transparent review process guided by Pawson et al.’s (2005) five key iterative steps in realist review; the Realist And Meta-narrative Evidence Syntheses—Evolving Standards quality and publication standards for realist reviews (Wong et al., 2013) and a range of protocols, methodological papers and reviews of and recommendations for practice including Booth et al. (2019); Dada et al. (2023); Duddy & Wong (2023); Rycroft-Malone et al. (2012); Saul et al. (2013), and Wong (2018). Also integral to this review is the establishment of an advisory panel to consult over the course of the review and collaboratively agree the finalised middle range programme theory. At this point, the advisory group has been established and Step 1 of the review has been completed. Step 2 is underway. Given that the review process is iterative, overlap across steps and backward forward movement is anticipated throughout the process.

Set up review advisory panel

Duddy and Wong (2023, p. 1) describe realist review as a “flexible, iterative and practical” approach to evidence synthesis that draws on the expertise of a variety of relevant stakeholders. Consequently, when undertaking a realist review a high level and variety of knowledge and expertise is required (Saul et al., 2013). In order to realise such expertise and knowledge, this review adopted a collaborative model incorporating an advisory panel (Davies et al., 2019; Shé et al., 2018). The composition of this advisory panel is overviewed in Table 3 with rational for inclusion of the specific advisors and summary of their expected roles. Also, a summary of the advisory panel meetings throughout Step 1 of the realist review is included in extended data.

Enabling active involvement of PPI advisors in realist review

In order to ensure the PPI advisors, especially those with communication impairments, engage actively and authentically on this advisory panel it is important to adapt a flexible and facilitative approach (McMenamin et al., 2022). Successfully negotiating complexity, such as realist theory development and refinement, with people with stroke acquired communication impairments, requires careful consideration and accommodations (Hersh et al., 2021). In this project the PhD candidate YF, is an experienced speech and language therapist who has worked with people with acquired communication difficulties for over 30 years. This according to Cascella & Aliotta (2014) renders her uniquely skilled in facilitating inclusive communicative interactions with people with acquired communication impairments. Annually, YF, RMcM and both PPI advisors with communication impairments co-deliver CPT to student speech and language therapists at the University of Galway. All are experienced in using supportive communication techniques and strategies to ensure active and reciprocal involvement during the training module. These communicative supports, tailored to the needs of each individual PPI advisor (one presenting with mild expressive aphasia and one with severe expressive aphasia and severe apraxia of speech), are being employed across this realist review to ensure understanding of the review process; clarity around the commitment and work required; active collaboration, and that there is shared understanding among the researchers and the PPI advisors. Ensuring there is comprehensive, shared understanding of the collaborative process aligns with Abrams and colleagues’ series of prompts for researchers and PPI contributors engaging in realist reviews (https://www.spcr.nihr.ac.uk/news/blog/the-role-or-not-of-patients-and-thepublic-in-realist-reviews), and with McMenamin et al. (2021, p. 17) “Top Tips” for researchers including people with aphasia in research. Examples of these tips and prompts include:

• Clarifying the point of the research and whom it will benefit

• Clarifying the PPI advisors expected time and work commitments

• Clarifying language demands e.g., how much reading would be required

• Building Relationships and ensuring there is equality and trust and people with aphasia are facilitated to engage actively and authentically

• Adopting an inclusive tone and ethos

• Identifying the most appropriate communication channels for use during the review

Specific examples of supportive strategy, employed in this review for PPI advisors with communication impairments, include:

• PPI Advisor 1 – Presenting information pictorially and in short written phrases (See sample in extended data); giving extra time to process information; supplementing conversation and verbal questions and instructions with written words and symbols; facilitating the use of a variety of non-verbal modes of communication including a communication folder, smart phone and tablet.

• PPI Advisor 2 - giving extra time to process verbal and written information, presenting verbal questions and instructions “one at a time”.

Time was also dedicated to the third advisor, the spouse of contributor 1, to ensure there was shared understanding of what engaging in the review entailed. She was provided with written information and engaged in discussion around the requirements of the project.

With regard to adopting a flexible approach, it was collaboratively agreed by YF and PPI advisors that meetings should be face to face and conducted separately with the the single advisor and the dyad (person with communication impairments and his significant other). This was due to their significantly different communication impairments and supportive requirements and to facilitate optimal use of everyone’s time.

Step 1

A. Clarify scope

The review questions were conceptualised by the content advisors and refined in consultation with the realist advisors. The questions aim to address shortcomings in the current knowledge base and facilitate realist enquiry. In order to determine how SH&SCPs learn to communicate optimally with people with stroke acquired communication impairments, we posed the following questions:

• What CPT interventions are used for SH&SCPs in higher education institutions?

• What are the desired, achievable outcomes of CPT in higher education institutions?

• For whom do these interventions work (or not) (e.g., which SH&SCPs and at what stage of their training; do they benefit people with stroke acquired communication difficulties participating as co-trainers or communication partners)

• What contexts (e.g., organisational, pedagogical, psychosocial) are enabling/inhibitory?

• How do these interventions work (what mechanisms are enabled in specific contexts to operationalise desired/undesired/unexpected outcomes)?

B. Articulate key theories to be explored

Concurrent with question refinement was initial programme theory (IPT) development. This is a fundamental, early step in a realist review (Pawson et al., 2004; Wong, 2018). Our IPTs set out how CPT for SH&SCPs is thought to work. They will be subject to testing and refinement in the subsequent synthesis process of this review. Figure 3 overviews the IPTs development process.

Figure 3. IPT development process.

Preliminary IPTs were developed by content advisors, guided by their insights into CPT, their knowledge of literature on the topic, and an additional literature scoping exercise (see extended data). YF formulated initial theories in the form of “If.., then…” statements. These were modified and refined iteratively over a five-month period in consultation with the other content advisors. This process generated nine “if.., then…” statements which were presented in context, mechanism, outcome configurations (CMOCs). In consultation with the realist advisors the CMOCs were developed further and amalgamated into a graphically presented, overarching IPT. This overarching graphic was additionally reviewed by a medical educationalist. Narrative and graphic IPTs were then presented to student advisors and PPI advisors for review and revision. Final revisions were agreed in consultation with content and realist advisors (see extended data). Narrative IPTs are presented in Table 4. Figure 4, graphically presents the amalgamated IPT, theorising how SH&SCPs can progress to being positive, productive communication partners, or not, and the reasoning and responses enacted along the way.

Figure 4. Graphic representation of integrated IPT for Stroke specific CPT for SH&SCPs.

Step 2 Search for evidence

This review aims to include a variety of articles and documents that address the review questions and can contribute to programme theory development and refinement (Davies et al., 2018; Luetsch et al., 2020). In line with the realist approach this search for relevant articles and documents will be iterative and conducted in collaboration with a research services librarian. The search progression will be responsive to emergent data and developing understanding and insight (Booth et al., 2019; The RAMESES Quality Standards for Realist Synthesis, The RAMESES Project (www.ramesesproject.org) ©2014). A concept-based search strategy, devised by YF in collaboration with the research services librarian and content experts, will direct an initial pilot search and data extraction (see Table 5). The search strategy will subsequently be refined iteratively and collaboratively with the advisory group. The searches will include peer reviewed journals, international best practice statements and clinical guidelines and conference proceedings. The following electronic data bases will be searched: Medline, EMBASE, CINAHL, APAPsycINFO and Web of Science. As outlined in our review protocol registered on PROSPERO (CRD42023418951), building from previous reviews (Simmons-Mackie et al., 2010; Simmons-Mackie et al., 2016; Tessier et al., 2020) searches of peer reviewed literature will be limited to the English language and by date – from January 2019 to the time of the review. Relevant studies in the previous reviews will be included for analysis. Supplementary searches will include hand searches of reference lists, and requests for unpublished studies/programmes from key authors. Search alerts will be in place to identify studies relevant for inclusion prior to final analysis.

Step 3 Select and appraise evidence

YF will perform the Title/Abstract screening and select documents consistent with preliminary inclusion criteria, that is, CPT for health and social care professionals or SH&SCPs addressing stroke acquired communication impairments. A random 20% selection will be reviewed by RMcM. Potential disagreements will be resolved through discussion and consensus of a third author as required. Endnote 20 will be the reference management system used and Covidence 2.0 systematic review software will be used to screen titles and abstracts.

In realist reviews, inclusion of data is determined by their ability to assist in the development and refinement of programme theory or theories. Hence, multiple and varied data sources are considered for inclusion (Wong, 2018). Applying quality scales or tools to such data sources risks excluding data that are essential for programme theory development. Therefore, in line with Price et al. (2021) in their realist review on remediating doctors’ performance to restore patient safety, formal quality appraisal tools will not be used in this review, on the basis that they are not sensitive to how data within papers contribute to programme theory development. Also, specific to CPT research, the application of quality scales has previously highlighted variable methodological quality of studies (Cherney et al., 2013; Simmons-Mackie et al., 2016). Simmons-Mackie et al. (2016) recommend that the purpose of the study must also be considered when assessing its value and Pawson et al. (2005, p. 24) recommend that when undertaking a realist review judgement around “fitness for purpose” needs to be made.

In accordance with The RAMESES Quality Standards for Realist Synthesis The RAMESES Project (www.ramesesproject.org) ©2014; Booth et al. (2013); Wong (2018), and Dada et al. (2023) any section of a document included in this review will be appraised for:

Adhering to the above standards and recommendations bespoke appraisal tools are being devised for this review. Key concepts are synopsised in Figure 5.

Figure 5. Synopsis of key concepts for appraisal of relevance, richness and rigour (adapted from Wong, 2018; Dada et al., 2022 and Dada et al., 2023).

Step 4 Extract the data

The RAMESES Quality Standards for Realist Synthesis (The RAMESES Project (www.ramesesproject.org) ©2014, p.7) advises that the data extraction process be continually refined over the course of the review in line with evolution of programme theory and focusing of review questions. Consequently, as Pawson et al. (2004, p. 23) had previously highlighted, data extraction in realist review is a process “without an exact equivalent”. A preliminary data extraction template has been devised and includes study, participant and intervention characteristics, outcome measures and outcomes, contexts, mechanisms of action and underpinning theories. For analogous purposes the domains of the substantive theoretical framework the TDF (Atkins et al., 2017) and COM-B elements of the BCW (Michie et al., 2011) are included in the extraction template. This template will be amended as indicated following the pilot searching, and iteratively thereafter in line with evolving insights and understanding. Once coded with respect to richness, relevance and rigour sufficiently rigorous data will be prioritised and assigned conceptual labels relating to CMOCs or components thereof. Less rigorous data will also be interrogated and undergo a triangulation process prior to contributing to programme theory refinement.

Step 5 Analyse and synthesise data

Data synthesis centres on programme theory refinement (Hunter et al., 2022; Pawson et al., 2005; Rycroft-Malone et al., 2012). Once thoroughly familiar with the extracted data sets, interpretations and judgements will be made around:

The following analytical process adapted from Pawson (2013) by Papoutsi et al. (2018, p. 16) will inform analysis and judgement throughout the synthesis process:

Matrices will be devised to facilitate this analysis. Analysis, judgement and synthesis will be a collaborative and iterative group process, see Figure 6. Throughout this process uncertainties will be resolved through discussion, debate and group consensus.

Figure 6. Group analysis, judgement and synthesis process.

Study status

As outlined in Figure 2, this realist review comprises the first step of this project. Step one of the review has been completed and step two is underway.

Plan for dissemination

This protocol will be submitted for publication to a peer-reviewed publishing platform. The findings of the completed review will be presented in a second article, adhering to the RAMESES publication standards for realist synthesis, and submitted for publication to a peer-reviewed journal. Also, following consultation with the advisory panel other relevant platforms including conferences, workshops, media platforms, and special interest groups will be explored for dissemination.

Ethical approval

Ethical approval is not required for this review. No data will be collected from the advisory panel.