Realist-informed synthesized narratives: A novel method for promoting a healthy speaking up culture in healthcare

Abstract

A goal in implementing health research is to change behavior and workplace culture. Realist methods are being used increasingly in healthcare research. While these methods provide context-based pragmatic recommendations, the data can be dense. Thus, making the transfer of knowledge into practice challenging for researchers and clinicians. This paper offers a novel approach to studying complex issues, communicating and motivating behavior change in healthcare using a realist approach to narrative analysis and producing synthesized narratives. A realist lens to research enables researchers to understand what works, with whom, and under what conditions. When used alone, realist methods can result in complex findings that can prove challenging to translate into practice. However, combining a realist approach with narrative analysis can enable a better understanding of the topic and promote practice change. This paper employs a case study on speaking up in healthcare to illustrate this novel method. This case study reports on a longitudinal interview study with a cohort of allied health new graduates to illustrate these methods and discuss the benefits and limitations of their application. We will provide researchers with clear steps to support replication. We argue that this method can aid the implementation of research findings across various contexts within and outside healthcare settings.

Keywords

Voice behavior speaking up new graduate allied health realist methods narrative analysis patient safety quality improvement

Background

Voice behavior is a complex, multifaceted phenomenon with multiple outcomes depending on the situation and people involved, making it a challenging behavior to adapt. Nuanced knowledge is required to understand what works, with whom, and in what situations to affect behavior change. A realist approach to research can be helpful in providing this level of detail and a lens on causality (Jagosh et al., 2022). However, a limitation of realist approaches is that while they help build theory, they can result in dense content, making knowledge transfer, and resulting behavior modification challenging. This paper outlines these challenges and discusses a novel methodological approach that combines realist and narrative methods to explore and illustrate speaking up in healthcare. We will focus on why realist approaches can clarify complex issues, why narratives are a powerful way to communicate information and stimulate change, and how combining realist and narrative methods can support the implementation of pragmatic research-informed recommendations. Within this paper, we use a longitudinal interview study with a cohort of new graduates as a case study to illustrate these methods and discuss the benefits and limitations of their application. We will provide researchers with clear steps to support replication. We argue that this method can aid the implementation of research findings across various contexts within and outside healthcare settings.

Definitions

Speaking up

Voice behavior is a broad term that describes a variety of workplace communications. This term can include prosocial voice behavior, patient advocacy voice, safety voice, speaking up, whistle-blowing, and unionized voice (Lee et al., 2020). In this paper, we will use the term “speaking up.” Speaking up is a critical interprofessional communication competency (Law and Chan, 2015). While speaking up may seem simple, how and why healthcare staff speak up or not is complex and poorly understood (Violato, 2022). We describe speaking up as informal, direct voice behavior that aims to improve a situation, is constructive, and change-orientated. This definition derives from the work of Van Dyne and LePine (1998), Liang et al. (2012), and Kolbe et al. (2012). Thus, speaking up may include highlighting an error, suggesting an improvement, or giving an opinion (Martinez et al., 2017; Schwappach and Gehring, 2014). This behavior may relate to patient or staff safety, quality care, and quality improvement. Effective speaking up reduces errors, improves patient safety, enhances quality improvement, and increases job satisfaction (Morrison, 2023; Weiss and Zacher, 2022; Wilkinson et al., 2020). This study focused on the speaking up experiences of allied health new graduates in healthcare.

Allied health new graduates

New graduates can be defined as healthcare workers in their first 2 years of practice following graduation (Dyess and Sherman, 2009). Allied health includes healthcare workers outside of nursing or medicine. They are a collective of professions representing a diverse workforce responsible for providing diagnostic, rehabilitative, therapeutic, technical, and support services to clients, whānau (family), and communities (Gilmour et al., 2014; Theisen and Sandau, 2013).

Speaking up in healthcare is complex

Speaking up is a complex phenomenon to understand, illustrate, and modify (Morrison, 2023). A scoping review by Friary et al. (2021) defined and described the breadth of speaking up in healthcare. This review (n = 75) included studies published between 2009 and 2019. The review highlighted how speaking up in healthcare continues to gain global interest, with publications growing. However, it noted that limited published research reports on the experiences of new graduates in allied health and few studies take a realist approach or explore changes in speaking up over time (Friary et al., 2021). Within the scoping review, 17 articles were identified describing intervention studies aiming to target improvements in speaking up, yet only four articles described improvements in speaking up secondary to the intervention. These studies often reported changes in knowledge, attitude, and confidence in speaking up, but did not report on a translation to behavior (Friary et al., 2021). This finding highlights the challenges researchers and healthcare leaders face when implementing research-based practice changes.

Given that speaking up in healthcare is a multifactorial, non-linear process, factors that facilitate or prevent speaking up may be social, cultural, and/or political in nature (Etchegaray et al., 2020; Umoren et al., 2022; Violato, 2022). Speaking up outcomes will differ depending on who is involved and the context. Without taking an in-depth perspective when exploring speaking up, researchers risk recommending ineffective interventions, as indicated in a review of speaking up interventions by Violato (2022). This review also highlighted the need for future studies to take a longitudinal perspective on speaking up and to gain an understanding of the causative mechanisms (Violato, 2022). The two-phase study reported on in this paper addresses these issues by viewing speaking up overtime using realist and narrative methods. Realist research methods provide one approach to developing an in-depth, nuanced view of speaking up behavior.

Realist research methods support an understanding of complex problems

Realist research is based on a philosophy of science, providing researchers with an alternative paradigm to positivism and constructivism (Maxwell, 2012). Realist research can be further defined as Critical Realism or Scientific Realism, in addition to other terms (Mukumbang et al., 2020, 2023). This paper has been influenced by Scientific Realism (Pawson and Tilley, 1997), a theoretical approach increasingly applied to healthcare, social sciences, and education program evaluations (Aunger et al., 2021; Davies et al., 2019; Jagosh et al., 2015; Pawson and Tilley, 1997; Williams et al., 2017). Realist philosophy allows the researcher to move away from the dualism of taking either a positivist or constructivist view of knowledge. Instead, it provides an opportunity to hold both physical knowledge (i.e. what we can count and measure) and mental knowledge (i.e. data about ideas, thoughts, and experiences) as equal (Emmel et al., 2018). Such a stance means that realist research can be a useful methodological approach when researching complex issues requiring pragmatic solutions, such as speaking up. Speaking up is a dynamic process and a multi-dimensional phenomenon to study. A realist lens invites the researcher to explore the underlying mechanisms involved in social events using retroduction and abduction—reasoning processes used by realist researchers to explore social events (Jagosh, 2020). These heuristics enable the researcher to look at how and why an outcome occurred rather than just identifying that it did occur or only describing the event. Realist methods will provide an understanding of the elements that work together to affect speaking up. This detail will support pragmatic recommendations for education and practice settings tasked with instilling a healthy speaking up culture.

To answer complex research questions, realist researchers are interested in working with data to understand context and causality (Maxwell, 2012; Pawson and Tilley, 1997). This data could be gathered through several methods, including interviews, focus groups, document reviews, and observations (Jagosh, 2020). In realist research, constructing context-mechanism-outcome configurations (CMOCs) can be used heuristically to support data analysis (Pawson and Tilley, 1997). Context refers to the setting in which phenomena are being explored. Causes can also be described as mechanisms—factors that generate or prevent change (Kjørstad and Solem, 2017). Understanding contexts that may trigger mechanisms, mechanisms contributing to or constraining change, and outcomes generated can support researchers in understanding complex social situations and identifying practical recommendations for change (Kjørstad and Solem, 2017). When looking at speaking up using a realist lens, the context includes elements in the background that have an impact on speaking up outcomes (e.g. team culture), mechanisms are resources that enable or prevent speaking up (e.g. information provided), and how people think about this (e.g. feeling of confidence influencing different outcomes). The outcomes are the intended and unintended effects of combining context and mechanisms, for example, speaking up effectively or not speaking up (Jagosh et al., 2015; Pawson and Tilley, 1997). This depth of exploration enables researchers to develop recommendations relevant to specific settings and situations, yet the theory-building component of this research approach creates portable theory that can be applied to different settings (Shearn et al., 2017; Westhorp, 2012).

The reasoning processes used to postulate mechanisms are retroduction and abduction (Jagosh, 2020). Realist research methods use iterative processes to build explanations for why outcomes occur in some settings and not others (Wong et al., 2016). This research process allows researchers to explore social problems in-depth, identify practical solutions, and identify the key factors for successful program implementation (Fletcher, 2017). Given these factors, realist research methods provide an apt approach for studying complex issues in different contexts.

Many studies exploring speaking up adopt a positivist approach and involve large-scale cross-sectional surveys (Friary et al., 2021). While this approach enables researchers to generate large volumes of data, address hypotheses, and identify patterns, it does not assist them in gleaning the underlying processes potentially causing the patterns (Wong et al., 2012). The research study drawn on in this paper was interested in understanding what causes or prevents new graduates’ speaking up behaviors in healthcare. By taking a realist perspective, we aimed to gain an understanding of the contexts and causal mechanisms that work together and result in new graduates speaking up effectively or not. We anticipated that this would support the development of theory to help explain this phenomenon while providing some pragmatic recommendations on improving speaking up behavior from the individual, healthcare team, and wider organizational perspective.

This study combines realist and narrative methods in the analysis and illustration of the findings. While realism and narrative analysis stem from different theoretical positions, some researchers argue that realist and qualitative methods can be combined, given that realists agree to a form of epistemological constructivism (Maxwell, 2012; Huberman and Miles, 2002). Further justification for the combination of these methods can be found within the Discussion.

Implementing change using realist-informed, synthesized narratives

A challenge for many researchers is how to implement research findings into practice. Narratives are a powerful way to translate and transmit knowledge to activate behavior change (Brooks et al., 2022; Riessman, 2008). They are an effective way to communicate complex information that is meaningful and memorable, and enhances learning (Greenhalgh, 2016; Greenhalgh and Hurwitz, 1999). Narratives are a tool commonly used in healthcare and other complex social settings to support quality improvement projects and promote behavior change (Greenhalgh et al., 2005; Perrier and Martin Ginis, 2018; Rose et al., 2016). Storytelling is an age-old approach to disseminating knowledge across generations (Polkinghorne, 1995). By connecting with the audience with humor, emotion, data, and detail that fosters connection, the audience can be motivated to remember content and engage in the details—all of which contribute to behavior change (Gallo, 2019). This paper outlines a research approach that combines realist and narrative methods from data collection, analysis, and illustration of findings, with the aim of affecting behavior change using an illustrative case study from a healthcare setting. The methods section details how we applied realist and narrative methods to enable replication for future studies. We propose that this combination of methods strengthens the findings of this study by translating the dense yet pragmatic recommendations gleaned from the application of realist methods into relatable and engaging narratives using narrative methods.

Methods

The case study we use to illustrate this approach of combining realist and narrative methods to produce causal synthesized narratives stems from a two-phase study exploring speaking up in allied health new graduates. The research questions were:

RQ1: How do new graduates in allied health experience speaking up in healthcare settings?

Phase 1 of this study explored RQ1 using a series of focus groups with new graduates and supervisors of new graduates. Using realist methods, an Initial Program Theory was generated about speaking up behavior in allied health new graduates (Friary et al., 2023a, 2023b). The decision to draw on narrative analysis methods was made during the analysis of the Phase 1 focus group study. During this study, it was apparent that participants shared their experiences using stories. They used narratives to illustrate their thoughts, opinions, and reasoning. It was also apparent that the stories shared provided insights into context, social relationships, and culture, all features of narrative analysis (Breheny et al., 2021).

The Initial Program Theory developed in Phase 1 was then developed, tested, and refined further throughout Phase 2 of the study, which answered the following questions:

RQ2: What mechanisms support and impact new graduates’ speaking up?

RQ3: How do speaking up behaviors change over a 1-year period?

RQ2 and RQ3 were answered using a longitudinal cohort study that employed semi-structured interviews with 10 allied health new graduates over 1 year, at three different time points. In this longitudinal study, we developed and applied realist-informed methods with narrative analysis to understand and illustrate causation (Maxwell, 2012). Informed by a realist approach, we developed synthesized narratives using data patterns across and within participants to create in-depth descriptions of different speaking up outcomes for new graduates in allied health.

This study has been reported using the Standards for Reporting Qualitative Research Ethics. The University of Auckland Human Participants Ethics Committee approved ethics (no: 024594).

Given that the findings from this study have been previously published (Friary et al., 2024), this paper will focus on the methods used to combine realist and narrative methods. Table 1 describes the steps taken from data collection to narrative construction.

Table 1.

Steps taken from data collection to synthesized narrative compilation.

Step	Description
1	Participants were interviewed using realist interviewing techniques in a semi-structured interview format. Questions were developed from the Initial Program Theory developed in Phase 1 of this research project. Phase 2 interview study was designed to test, refute, and refine.
2	Interview audio recorded and transcribed.
3	Notes taken by the researcher during the interview related to insights.
4	The participant was emailed a summary of the interviewer’s insights to check for accuracy and to see if they had anything to add.
5	The researcher immersed themself in the transcription in a non-analytical way and re-read it several times.
6	The transcription was re-read, and text that illustrated causal insights related to speaking up was highlighted with comments added in the margins.
7	Chunks of text were transferred into a matrix and extracted and elicited into CMOCs (context-mechanism-outcome configurations). This is an iterative process completed from the position of the lead researcher.
8	Quotes from the transcription were added to illustrate each CMOC.
9	Using these CMOCs, the researcher developed “If . . . then . . . ” statements and added them into another column within this CMOC coding matrix. This process was completed for all 10 round 1 interviews. This step shows the early stages of the development of the IPT into the Mid-range Theory.
10	Data was analyzed immediately following the interview so that in the next round of interviews, the researcher could further test, refute, and refine the developing Mid-range Theory. This became a more deductive, theory-testing interview style—a particular realist interviewing style often described as a teacher-learner approach, where the interviewer and interviewee are building theory together.
11	For the second and third interviews for each participant, steps 1–11 were completed and added to each individual participant’s CMOC coding matrix.
12	Patterns of regularity across all data sets were reviewed and illustrated using an online Miro board (Miro, 2022) to explore causal networks. A causal network started to develop for four different outcomes related to speaking up.
13	To add more ontological depth to these outcomes stated in 12, narrative analysis was used to re-story the analysis as related to each outcome to outline the context and causal mechanisms triggering these outcomes. This could also be described as a re-configuration of the Mid-range Theory into narratives illustrating causation. This was a backward mapping process whereby the researcher started with the outcomes and worked backwards within and across all participants to identify the factors related to context and mechanisms responsible for influencing different outcomes. The outcome stories could also be described as archetypes related to speaking up, that is, the new graduates who speak up effectively. These archetypes are a synthesis across all 10 participant interviews (i.e. total of 30 interviews). The narratives are representations of reality, as shared by all participants, and include their combined experiences of speaking up in healthcare using real accounts and examples. They are, therefore, combined explanatory accounts grounded in realist ontology and constructivist epistemology.
15	The synthesized narratives were written using Labov’s framework to write a compelling story. The narratives were written using large chunks of raw data from the interview transcripts to illustrate different stages of the story.
16	These four synthesized narratives collectively form the Mid-range Theory on allied health new graduates’ speaking up experiences.

NB: At various points throughout this process, the analysis was reviewed with other research team members for accuracy and clarity. Any recommendations were incorporated into the development of the theory.

Longitudinal cohort study

Participants engaged in semi-structured interviews three times over 1 year at approximately 4-month intervals. In total, 30 interviews were conducted. Consistent with narrative analysis, a small sample of new graduates was recruited to ensure rich data and a deep understanding of their speaking up experiences (Polkinghorne, 1995).

The first author, skilled in semi-structured interviewing, conducted the interviews. All interviews were conducted via videoconferencing, audio recorded, transcribed, and de-identified. The first author then checked the transcriptions for accuracy. To ensure research rigor, a summary of the interpretation of the interview was shared with all participants to check for credibility.

Combining methods

A map presenting the combination of realist and narrative analysis methods is illustrated in Figure 1.

Figure 1.

Figure showing the combination of methods.

Within a realist interview (Appendix 1) using a teacher-learner interviewing style (Manzano, 2016; Mukumbang et al., 2020, 2023), participants were asked to provide recent examples of when they spoke up or remained silent. Questions were used to elicit details about the factors activating or inhibiting speaking up and the effectiveness of this communication exchange. This dynamic, iterative questioning process allowed the interviewer to work with participants to refine the Initial Program Theory (developed in Phase 1), thereby collectively forming a Mid-range Theory. This interviewing technique is a key feature of realist research. In addition to this realist technique, the interview schedule was influenced by a narrative approach, as we invited the participants to share their experiences using a long story form. In line with narrative interviewing techniques, we completed multiple interviews, often recommended in narrative inquiry literature. Questions were open and explored the participants’ experiences of events with depth and detail (Lyons and McAllister, 2019; Lyons et al., 2021; Patton, 2002; Riessman, 2008).

Interview transcripts were analyzed and coded using a realist coding technique—context-mechanism-outcome configurations (CMOCs) (Jagosh et al., 2022; Pawson and Tilley, 1997). This technique involved the researcher becoming familiar with the transcripts through multiple readings and then reviewing the transcripts to look for features of causation. For example, these features could include resources available to new graduates, such as being specifically asked for feedback in a team meeting, or their feeling state, such as the feeling of empowerment. From this, the researcher coded context, mechanism, and outcome elements. These coded elements formed the CMOCs. The CMOCs were then collated into a CMOC matrix for each participant’s interview, which was further developed into an overall individual Initial Program Theory (IPT) for each of the three individual interviews (Jagosh et al., 2022; Pawson and Tilley, 1997). An IPT can be expressed as a causal statement (e.g. “if . . . then . . . ”) describing why a particular set of resources results in specific outcomes caused by a relationship between context and mechanism elements (Jagosh et al., 2022). An IPT example can be found in Figure 2. Further analysis sought patterns of regularity within and across participants. These patterns were collated and mapped into four speaking up outcomes. We also introduced a backward mapping process from the four common speaking-up outcomes to identify and collate mechanisms and contextual factors related to these outcomes across all participants (Beutel et al., 2023). In this way, the IPTs were iteratively refined into a Mid-range Theory over the 12-month data collection and analysis period. A Mid-range Theory is described by Jagosh (2019) as a theory that is not so abstract as to be disconnected from real-life events and is developed by testing, refining, and refuting the IPT to a deeper level of abstraction. The Mid-range Theory articulated four different hypotheses explaining why speaking up outcomes occurred for the new graduates in this study. A Mid-range Theory is sufficiently abstract to be taken and applied to different contexts (Waldron et al., 2020). Through a logic process of abduction, a Mid-range Theory enables social scientists to highlight the context-specific mechanisms related to different outcomes while enabling analytical generalization of the findings to similar contexts (Mukumbang and Wong, 2025).

Figure 2.

Exemplar illustrating key steps in analysis from interview to development of synthesized narratives.

Large chunks of text from the transcriptions were identified and used to illustrate context features and mechanisms working together to contribute to different speaking up outcomes. This information was then expanded to construct four synthesized narratives, as detailed in Figure 2.

Using the CMOCs previously coded and the outcome matrix, we developed synthesized narratives for each speaking up outcome. We aimed to articulate specific factors related to context and mechanisms contributing to speaking up outcomes. A re-storying process was adopted using Labov’s narrative framework (Yardley et al., 2020). Large sections of text from all transcribed interviews were embedded throughout the synthesized narratives to ensure representation of all new graduates’ experiences. While Labov’s framework is typically used to analyze narratives, we used it in a novel way to construct narratives. Table 1, Figures 1 and 2 illustrate this process.

Trustworthiness

Throughout the study detailed in this paper, the trustworthiness of the findings was maintained through peer review of coding and the outcome matrix by more experienced realist and narrative inquiry researchers. Four research team members reviewed the narratives for quality and authenticity, and an audit trail was kept throughout all stages of the study, including reflective writing and memos related to thoughts about the data (Lyons et al., 2021; Patton, 2002, 2014). This occurred at all stages of the study.

Reflexivity

A team of researchers from various professional and research backgrounds collaborated in this study, including speech-language therapists, an audiologist, an educational researcher, and a rehabilitation lecturer. The lead researcher is a clinician and educator teaching healthcare students at hospitals and universities. While no participants were current students or supervisees of the research team, one was a previous student of the lead researcher. These dual roles and connections with some of the participants could have impacted what they shared during the focus groups, that is, resulting in them being more guarded (or less guarded) with the information they provided. Additionally, the professional roles and experiences of the lead researcher have influenced all aspects of the study, from the determination of the research questions to the interpretation of the findings.

Reflexivity was maintained throughout this study through the research team’s ongoing discussion and critique of assumptions (Huberman and Miles, 2002; Lingard and Watling, 2021; Soedirgo and Glas, 2020; Wilson et al., 2022). The lead researcher kept a reflexive diary throughout the study and returned to the transcriptions regularly to remain close to the participants’ stories and reflect on any biases and assumptions concerning the data.

The new graduates in the cohort study reported that they learned about the speaking up context when asked about it during the interviews. The well-known Hawthorne Effect occurs when participants’ behaviors change when they know they are being observed (McCambridge et al., 2014). While we were unable to control for this effect, discussions with the participants highlighted their thoughts on whether their speaking up behaviors changed with speaking about speaking up. Several participants said they did not think this occurred and reported that their typical communication style was to speak up as needed. Others felt that they had become more aware of this type of communication and reflected on this between our meetings. While they did not think this necessarily aided them in speaking up more effectively, they did think about speaking up more frequently. This finding highlights the potential benefits of targeted training on speaking up, the need to raise awareness about speaking up across healthcare organizations and the benefits of bringing speaking up experiences and practice into supervision.

Credibility and confirmability

Credibility was attained throughout this study in several ways. Careful attention was given to the alignment between the research questions, the research paradigm, methods for data collection, and analysis methods. Justification of this alignment is discussed in this paper. A summary of each individual interview was shared with each participant to ensure we had an accurate interpretation and depiction of their experiences. In addition, throughout the Phase 2 interviews, participants were asked to reflect on factors that may have influenced their responses, such as the wider socio-cultural events and the experience of reflecting on speaking up with the interviewer. In addition to a comprehensive and continuous audit trail, we maintained clear links between the data and analysis using thick descriptions and quotes. A final feature of study credibility is the prolonged engagement that we had with the participants, given that they were interviewed three times over a 1-year period and the immersion and continual return to the transcriptions and data memos throughout the analysis and write-up phases (Lyons et al., 2021; Patton, 2002, 2014).

Transferability

This study has achieved transferability through purposeful sampling and clearly detailing the context in which it was completed. These factors support the transferability of findings and replication to other settings and populations (Enworo, 2023).

Findings

Combining realist and narrative approaches in Phase 2 of this study allowed the research team to develop synthesized narratives around four different speaking up outcomes from all participant data. This approach enabled us to combine different factors related to the contexts and mechanisms needed for different speaking up outcomes to occur in this new graduate cohort. We believe this combined approach more clearly illustrates and communicates complex, multi-factorial data, and summarizes the contextual and mechanistic features required for effective speaking up to occur.

Realist methods enabled development, testing, refining, and refuting IPTs to construct a Mid-range Theory that outlines four speaking-up outcomes (Figure 3). Narrative methods enabled the development of causal explanations for each speaking up outcome, evident in the stories shared by the participants. These were then further developed into four different synthesized narratives using a combination of narrative analysis and realist methods. These synthesized narratives highlight the elements related to context and causation that influence the four different speaking up outcomes. These are—(1) effective speaking up, (2) speaking up with unmet expectations, (3) withholding voice, and (4) speaking up less over time. See Friary et al. (2024) for a more detailed description of the synthesized narratives. The datasets used and/or analyzed in this study are available from the corresponding author upon reasonable request.

Figure 3.

Figure illustrating the mid-range theory on new graduates’ speaking up in healthcare.

In summary, the four different speaking up outcomes are related to:

Effective speaking up

This outcome occurred for several new graduates when they were engaged in relational and reflective supervision and interprofessional collaborative practice and had previous positive speaking up experience, in addition to several other elements noted in Figure 3. This resulted in several mechanisms activating, including feeling valued, empowered, and respected, a sense of belonging, and being committed to improving their speaking up skills.

Speaking up with unmet expectations

These graduates experienced a high workload and a fast pace. They described siloed workplaces and practices, a low-trust relationship with their supervisor, and limited orientation. Yet, they expressed that they valued ethical and quality practice and their role as patient advocates. This led to a mismatch in expectations and resulted in them having no sense of belonging and not feeling respected in their role. They also reported negative speaking-up experiences that contributed to increasing uncertainty about the value of speaking up.

Withholding voice

When new graduates received a limited orientation to the team and setting and then had a high workload within a team with a high risk of burnout, withholding voice was evident as an outcome. They expressed a lack of confidence in their technical knowledge and reported a team culture of modeling withholding voice. This resulted in the new graduates’ poor understanding of their role and low trust in their team. However, they also felt they needed to fit into the team and expressed a desire not to get anything wrong.

Speak up less overtime

The longitudinal study design provided 1 year of data related to speaking up. This perspective highlighted that many new graduates reported speaking up less over time. This occurred when new graduates experienced their previous suggestions not being implemented, leaders being unavailable, a team culture of withholding voice, pressure to conform, and increased workload. These contextual factors made new graduates feel apathetic, overwhelmed, and unsupported. They also demonstrated a lack of trust in leadership, disengagement from the organization, and reduced job satisfaction.

Discussion

Researchers and practitioners in healthcare face significant challenges in researching, illustrating, and implementing safety and quality improvement initiatives due to the complexity of healthcare systems. Two challenges we encountered within this study are discussed below: (1) justifying the use of various methods to develop pragmatic solutions to complex issues and (2) how to develop methodological and dissemination solutions to optimize the potential for behavior change. Finally, we discuss the benefit of a diverse research team in supporting and interrogating methodological developments.

Combining realist and narrative methods when studying complex issues

A key challenge for researchers is articulating research findings in ways that engage the audience and activate behavior change (Wailling et al., 2021). A common critique of speaking up intervention programs is that they are not research-based or specific to the contexts under investigation and do not promote the translation of new knowledge into practice (Morrison, 2023; Violato, 2022). Given the complexity of healthcare settings and the distinctiveness of different contexts concerning the people, the culture, and the broader socio-political climate, healthcare issues such as withholding voice or ineffective speaking up are wicked problems (Dixon-Woods, 2019; Dixon-Woods et al., 2022), requiring specialist-targeted solutions. In this paper’s case study, narrative and realist methods were combined at all study stages. From interviewing, analysis, and interpretation to communication of findings. A key benefit of employing realist methods to tackle complex issues is the attainment of analytic depth, resulting in the theoretical portability of findings into other settings and populations (Shearn et al., 2017; Westhorp, 2012).

Additionally, narrative methods elicit participant stories, drawing out cognitive, emotional, and socio-cultural factors (Polkinghorne, 1995; Riessman, 1993, 2008). We argue that by combining these methods, we have identified pragmatic and applicable solutions to a complex issue, such as speaking up in healthcare—while also potentially promoting audience engagement in the findings by facilitating connection with characters in the narratives and, therefore, better engagement with the knowledge being shared. Using realist methods alone can result in dense and extensive findings, making new knowledge challenging to connect with and implement. However, combining these techniques resulted in synthesized stories of causality, specifying which elements related to context and mechanism need to be activated to trigger effective speaking up outcomes in healthcare.

Exploring research topics by combining methods from different methodological approaches enables the uncovering of novel and nuanced discoveries. Speaking up is frequently researched using positivist analysis methods and large data sets (Friary et al., 2021). Additionally, realist studies frequently source data from mixed methods (Jagosh et al., 2015, 2022). Instead, this paper reports on a longitudinal study with 10 participants and 30 individual interviews. While we had fewer participants than large-scale survey studies, the interviews produced an extensive data set for a qualitative study.

Using synthesized narratives in implementing theory to practice

This paper describes a research approach combining realist and narrative methods to tell causality stories (Maxwell, 2012). Some researchers taking a purist approach to research argue that approaches stemming from different philosophical positions cannot be combined (Buetow, 2007; Guba and Lincoln, 1994; Slevin, 2010). This argument relates to differences in how researchers from different paradigmatic perspectives understand ontology and epistemology and how these constructs influence research. For example, researchers influenced by Positivist philosophers such as French Philosopher Auguste Comte or Post-Positivist philosophers such as Karl Popper, who believe reality is present and independent of our awareness of it and that knowledge can be measured, value research approaches that gather large data sets and complete tests such as hypothesis testing and cross-sectional analyses (Bradbury-Jones et al., 2017; Buetow, 2007; Thomas, 2006). In comparison, constructivists assert that knowledge is socially constructed (Crotty, 1998; Patton, 2002). When investigating people, outcomes will be different in different contexts—meaning that reality is individually created and will differ depending on the individual’s perspective (Crotty, 1998; Patton, 2002). In opposition to a purist position that argues against combining methods (Guba and Lincoln, 1994; Slevin, 2010), methodological pragmatists such as Maxwell (2012) and Patton (2001) argue that research methods are not inextricably linked to specific theoretical positions and that methods stemming from different philosophical perspectives can be combined based on pragmatic utility. We also argue that researchers can combine realist and narrative approaches, given that realism is method-agnostic and constructivism is broad and varied. This position has also been argued by Maxwell (2012) and Breheny et al. (2021). Maxwell (2012) outlines how several different constructivist research approaches can be used within a realist framework. He outlines ways researchers can use narrative analysis to identify causal connections and re-story data into different profiles or vignettes to represent the data. This perspective is also supported by Breheny et al. (2021), who took a critical realist position to explore racism using focus groups and narratives. Within this study, they discuss how individual short stories can also be described in narrative terms as a collection of scenes that can then be cross-analyzed between and across participants for patterns. There are a few published examples of combining realist and narrative methods, such as this study by Breheny et al. (2021). However, the creation of realist-informed synthesized narratives to communicate knowledge and stimulate behavior change in our case study is novel. By combining these approaches in this illustrative case on speaking up, we employed a dynamic inquiry process through the realist interviewing technique to interact with the individuals’ narratives and create shared meaning and new theory (Daiute, 2013; Jagosh, 2020; Jagosh et al., 2015). By synthesizing these short stories on speaking up using realist methods, narrative analysis enabled us to transform the collective experiences of the participants into an illustrative, cultural product—narratives (Daiute, 2013).

Narratives provide educators with an effective educational tool (Brooks et al., 2022; Gallo, 2019; Riessman, 2008). We believe these synthesized narratives have the potential to promote a positive speaking up culture at multiple levels within the healthcare system—the individual, team, and organizational levels. These narratives have already been utilized at the individual level during workshops involving pre-registration students and new graduates focusing on entry-to-practice competencies. Anecdotally, the participants reported that the narratives were memorable and engaging as they involved the voices of people they could relate to. At the team level, the vignettes could be used as an engaging, accessible way to complete a team self-audit of their team culture and collectively explore ways that they could improve speaking up culture within their team. To promote a healthy speaking up culture in healthcare at an organizational level, they could be embedded into the educational curriculum of a series of workshops targeting healthcare leaders tasked to improve safety and quality culture within their workplaces. Alternatively, the narratives could be used as a part of an organizational-wide campaign promoting effective speaking up across a whole organization through educational posters and online content introducing the different vignettes to engage staff, provide information, and promote behavior change.

Benefits of an interdisciplinary research team

The decision to combine realist methods and narrative analysis was organic. The original research plan was to employ mixed methods using a realist theoretical perspective. However, at the beginning of the new graduate cohort interviews, it became apparent that the new graduates’ data would be better analyzed and represented using a combination of realist methods and narrative analysis. The new graduates were naturally sharing speaking up stories about their experiences, and the stories had common themes and outcomes across and within the participants, which could be portrayed effectively in narrative form. The decision to adapt methods was made possible by having a research team comprising researchers from different disciplines and methodological backgrounds. This mix of knowledge and experience enabled the research team to engage in rich methodological debate around the pros and cons of this shift and determine a rigorous research process. While interdisciplinary research teams can bring challenges (Trussell et al., 2017), an overt process of reflexivity combined with negotiation and trust created a cohesive research dynamic.

Limitations

The novel method employed in this case study analyzed across and within participant stories to create four causal narratives. While combining the narratives of 10 participants may dilute the authenticity of their individual experiences by using a realist approach, we have completed a backward mapping process from the outcomes to the different contexts and mechanisms. This has resulted in the narratives being a combination of individuals’ experiences grouped and linked to the related outcome. By presenting the findings as synthesized narratives, developed using realist and narrative methods, we can clearly illustrate the interaction between the contexts and the mechanisms for the four different speaking up outcomes, which provides rich data related to each outcome. This technique clearly illustrates the practical steps education and healthcare institutions need to take to create a healthy speaking up culture.

Conclusion

A common problem for researchers and clinicians is communicating research recommendations in a way that promotes behavior change. This paper addresses a gap in published research by outlining a novel approach to studying complex issues using a realist approach to narrative analysis and producing synthesized narratives. A realist lens to research enables researchers to understand what works, with whom, and under what conditions. When used alone, realist methods can result in complex, dense findings that can prove challenging to translate into practice. However, combining a realist approach with narrative analysis has enabled us to understand the topic of speaking up in allied health new graduates and portray these findings in a way that we anticipate could motivate behavior change. Stories can be memorable and engage people. When used in healthcare, they can motivate behavior change. Therefore, we anticipate that the combination of narrative analysis with a realist lens could promote the implementation of this study’s findings into healthcare practice and be applied to other complex socio-political settings.

Footnotes

Appendix 1 Acknowledgements

I wish to acknowledge everyone who has participated in this study and shared their experiences, which have informed this manuscript.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Ethical approval

This study has been reported using the Standards for Reporting Qualitative Research Ethics. The University of Auckland Human Participants Ethics Committee approved ethics (no: 024594).

ORCID iD

Philippa M Friary

Author biographies

Philippa M Friary and speech-language therapist, is the Director of Clinical Education for the Speech Science programme at The University of Auckland, New Zealand.

Rachelle Martin is a Physiotherapist, works as a Lecturer with the Rehabilitation Teaching and Research Unit (RTRU) in the Department of Medicine at the University of Otago, Wellington, New Zealand.

Suzanne C Purdy (Te Rarawa, NgāiTakoto) is a Professor and Audiologist, previous Head of School of Psychology. Previously (2003-2018) and was Head of Speech Science in the School of Psychology, Auckland, New Zealand.

Lindy McAllister is a Emeritus Professor of Work Integrated Learning in the Work Integrated Learning (WIL) Unit of the Faculty of Health Sciences, University of Sydney, Australia.

Mark Barrow is an Associate Prof, was appointed as Dean of the Faculty of Education and Social Work in 2018, University of Auckland, New Zealand.

References

Aunger

Millar

Greenhalgh

, et al. (2021) Why do some inter-organisational collaborations in healthcare work when others do not? A realist review. Systematic Reviews 10: 1–22.

Beutel

Tangen

Carrington

(2023) Backward mapping from outcomes: Using a realist evaluation lens to evaluate an international aid program. Evaluation and Program Planning 97: 102201.

Bradbury-Jones

Breckenridge

Clark

, et al. (2017) The state of qualitative research in health and social science literature: A focused mapping review and synthesis. International Journal of Social Research Methodology 20(6): 627–645.

Breheny

Stevenson

Bécares

, et al. (2021) Tabulating experiences of racism and racial discrimination across the life course. Stigma and Health 6(1): 43.

Brooks

Zimmermann

Lang

, et al. (2022) A framework to guide storytelling as a knowledge translation intervention for health-promoting behaviour change. Implementation Science Communications 3(1): 35.

Buetow

(2007) Health Research Methods: A Tabular Presentation. New York, NY: Nova Science Publishers.

Crotty

(1998) The Foundations of Social Research: Meaning and Perspective in the Research Process. Thousand Oaks, CA: Sage publications.

Daiute

(2013) Narrative Inquiry: A Dynamic Approach. Thousand Oaks, CA: Sage Publications.

Davies

Fattori

O’Donnell

, et al. (2019) What are the mechanisms that support healthcare professionals to adopt assisted decision-making practice? A rapid realist review. BMC Health Services Research 19: 1–14.

10.

Dixon-Woods

(2019) How to improve healthcare improvement: An essay by Mary Dixon-woods. BMJ 367: l5514.

11.

Dixon-Woods

Aveling

Campbell

, et al. (2022) What counts as a voiceable concern in decisions about speaking out in hospitals: A qualitative study. Journal of Health Services Research & Policy 27(2): 88–95.

12.

Dyess

Sherman

(2009) The first year of practice: New graduate nurses’ transition and learning needs. Journal of Continuing Education in Nursing 40(9): 403.

13.

Emmel

Greenhalgh

Manzano

, et al (2018) Doing Realist Research. Thousand Oaks, CA: Sage.

14.

Enworo

(2023) Application of Guba and Lincoln’s parallel criteria to assess trustworthiness of qualitative research on indigenous social protection systems. Qualitative Research Journal 23(4): 372–384.

15.

Etchegaray

Ottosen

Dancsak

, et al. (2020) Barriers to speaking up about patient safety concerns. Journal of Patient Safety 16(4): 230–234.

16.

Fletcher

(2017) Applying critical realism in qualitative research: Methodology meets method. International Journal of Social Research Methodology 20(2): 181–194.

17.

Friary

P. M.

Purdy Prof

S. C.

McAllister

Martin

(2023a) Speaking up in healthcare: An exploration of the allied health new graduate workforce. Internet Journal of Allied Health Sciences and Practice 21(1): 5.

18.

Friary

P. M.

Purdy

S. C.

McAllister

Barrow

Martin

(2023b) A realist informed study on speaking up in Healthcare–Supervisors’ perceptions of new graduates speaking up. Internet Journal of Allied Health Sciences and Practice 21(2): 22.

19.

Friary

Purdy

S. C.

McAllister

Barrow

(2021) Voice behavior in healthcare: A scoping review of the study of voice behavior in healthcare workers. Journal of Allied Health 50(3): 242–260.

20.

Gallo

(2019) Storytelling to inspire, educate, and engage. American Journal of Health Promotion 33(3): 469–472.

21.

Gilmour

ASM

Jones

Cowpe

, et al. (2014) Communication and professionalism skills of a new graduate: The expectations and experiences of dental foundation trainers. European Journal of Dental Education 18(4): 195–202.

22.

Greenhalgh

(2016) Cultural Contexts of Health: The Use of Narrative Research in the Health Sector. Geneva: World Health Organization. Regional Office for Europe.

23.

Greenhalgh

Hurwitz

(1999) Why study narrative? BMJ 318(7175): 48–50.

24.

Greenhalgh

Russell

Swinglehurst

(2005) Narrative methods in quality improvement research. BMJ Quality & Safety 14(6): 443–449.

25.

Guba

Lincoln

(1994) Competing paradigms in qualitative research. Handbook of Qualitative Research 2(163–194): 105.

26.

Huberman

Miles

(2002) The Qualitative Researcher’s Companion. Thousand Oaks, CA: Sage.

27.

Jagosh

(2019) Realist synthesis for public health: Building an ontologically deep understanding of how programs work, for whom, and in which contexts. Annual Review of Public Health 40: 361–372.

28.

Jagosh

(2020) Retroductive theorizing in Pawson and Tilley’s applied scientific realism. Journal of Critical Realism 19(2): 121–130.

29.

Jagosh

Bush

Salsberg

, et al. (2015) A realist evaluation of community-based participatory research: Partnership synergy, trust building and related ripple effects. BMC Public Health 15(1): 1–11.

30.

Jagosh

Stott

Halls

, et al. (2022) Benefits of realist evaluation for rapidly changing health service delivery. BMJ Open 12(7): e060347.

31.

Kjørstad

Solem

(2017) Theory and practice as a dynamic relation. In: Kjørstad

Solem

(eds) Critical Realism for Welfare Professions. London: Routledge, pp.145–151.

32.

Kolbe

Burtscher

Wacker

, et al. (2012) Speaking up is related to better team performance in simulated anesthesia inductions: An observational study. Anesthesia & Analgesia 115(5): 1099–1108.

33.

Law

Chan

(2015) The experience of learning to speak up: A narrative inquiry on newly graduated registered nurses. Journal of Clinical Nursing 24(13–14): 1837–1848.

34.

Lee

Phan

(2020) Prosocial Advocacy Voice in Healthcare: Implications for Human Resource Management. United Kingdom: Emerald Publishing Limited, pp.181–221.

35.

Liang

Farh

(2012) Psychological antecedents of promotive and prohibitive voice: A two-wave examination. Academy of Management Journal 55(1): 71–92.

36.

Lingard

Watling

(2021) Story, Not Study: 30 Brief Lessons to Inspire Health Researchers as Writers. Cham: Springer.

37.

Lyons

McAllister

(2019) Qualitative Research in Communication Disorders: An Introduction for Students and Clinicians. Havant: J & R Press Limited.

38.

Lyons

McAllister

Carroll

, et al. (2021) Diving Deep Into Qualitative Data Analysis in Communication Disorders Research. Havant: J&R Press Limited.

39.

McCambridge

Witton

Elbourne

D. R.

(2014). Systematic review of the Hawthorne effect: new concepts are needed to study research participation effects. Journal of clinical epidemiology, 67(3), 267–277.

40.

Manzano

(2016) The craft of interviewing in realist evaluation. Evaluation 22(3): 342–360.

41.

Martinez

Lehmann

Thomas

, et al. (2017) Speaking up about traditional and professionalism-related patient safety threats: A national survey of interns and residents. BMJ Quality & Safety 26(11): 869–880.

42.

Maxwell

(2012) A Realist Approach for Qualitative Research. Thousand Oaks, CA: Sage.

43.

Miro (2022) Miro Online Whiteboard (no version provided). RealTimeBoard, Inc. Available at: https://www.miro.com https://www.miro.com (accessed 2 September 2024).

44.

Morrison

(2023) Employee voice and silence: Taking stock a decade later. Annual Review of Organizational Psychology and Organizational Behavior 10: 79–107.

45.

Mukumbang

De Souza

Eastwood

(2023) The contributions of scientific realism and critical realism to realist evaluation. Journal of Critical Realism 22(3): 504–524.

46.

Mukumbang

Marchal

Van Belle

, et al. (2020) Using the realist interview approach to maintain theoretical awareness in realist studies. Qualitative Research 20(4): 485–515.

47.

Mukumbang

Wong

(2025) Mechanism-based middle-range theories: Using realist syntheses to reconcile specificity to context and generalizability. Qualitative Health Research. Epub ahead of print 12 March 2025. DOI: 10.1177/10497323251316401.

48.

Patton

(2002) Two decades of developments in qualitative inquiry: A personal, experiential perspective. Qualitative Social Work 1(3): 261–283.

49.

Patton

(2014) Qualitative Research & Evaluation Methods: Integrating Theory and Practice. Thousand Oaks, CA: Sage publications.

50.

Pawson

Tilley

(1997) An introduction to scientific realist evaluation. In: Chelimsky

Shadish

(eds) Evaluation for the 21st Century: A Handbook. Thousand Oaks, CA: Sage publications, pp.405–418.

51.

Perrier

Martin Ginis

(2018) Changing health-promoting behaviours through narrative interventions: A systematic review. Journal of Health Psychology 23(11): 1499–1517.

52.

Polkinghorne

(1995) Narrative configuration in qualitative analysis. International Journal of Qualitative Studies in Education 8(1): 5–23.

53.

Riessman

(1993) Doing narrative analysis. In: Narrative Analysis. Thousand Oaks, CA: Sage.

54.

Riessman

(2008) Narrative Methods for the Human Sciences. Thousand Oaks, CA: Sage.

55.

Rose

Chakraborty

Mason-Lai

, et al. (2016) The storied mind: A meta-narrative review exploring the capacity of stories to foster humanism in health care. Journal of Hospital Administration 5(1): 52–61.

56.

Schwappach

Gehring

(2014) ‘Saying it without words’: A qualitative study of oncology staff’s experiences with speaking up about safety concerns. BMJ Open 4(5): e004740.

57.

Shearn

Allmark

Piercy

, et al. (2017) Building realist program theory for large complex and messy interventions. International Journal of Qualitative Methods 16(1): 1609406917741796.

58.

Slevin

(2010) Approaches to healthcare research. In: Healthcare Research: A Textbook for Students and Practitioners, pp.11–38. Hoboken: Wiley-Blackwell.

59.

Soedirgo

Glas

(2020) Toward active reflexivity: Positionality and practice in the production of knowledge. PS: Political Science & Politics 53(3): 527–531.

60.

Theisen

Sandau

(2013) Competency of new graduate nurses: A review of their weaknesses and strategies for success. Journal of Continuing Education in Nursing 44(9): 406.

61.

Thomas

(2006) General medical practitioners need to be aware of the theories on which our work depend. The Annals of Family Medicine 4(5): 450–454.

62.

Trussell

Paterson

Hebblethwaite

, et al. (2017) Negotiating the complexities and risks of interdisciplinary qualitative research. International Journal of Qualitative Methods 16(1): 1609406917711351.

63.

Umoren

Kim

Gray

, et al. (2022) Interprofessional model on speaking up behaviour in healthcare professionals: A qualitative study. BMJ Leader 6(1): 15–19.

64.

Van Dyne

LePine

(1998) Helping and voice extra-role behaviors: Evidence of construct and predictive validity. Academy of Management Journal 41(1): 108–119.

65.

Violato

(2022) A state-of-the-art review of speaking up in healthcare. Advances in Health Sciences Education 27(4): 1177–1194.

66.

Wailling

Long

Stolarek

(2021) The context and influences of safety: An exploration of social networks, actions, and behaviour. International Journal for Quality in Health Care 33(1): 157.

67.

Waldron

Carr

McMullen

, et al. (2020) Development of a program theory for shared decision-making: A realist synthesis. BMC Health Services Research 20: 1–17.

68.

Weiss

Zacher

(2022) Why and when does voice lead to increased job engagement? The role of perceived voice appreciation and emotional stability. Journal of Vocational Behavior 132: 103662.

69.

Westhorp

(2012) Using complexity-consistent theory for evaluating complex systems. Evaluation 18(4): 405–420.

70.

Wilkinson

Barry

Morrison

(2020) Toward an integration of research on employee voice. Human Resource Management Review 30(1): 100677.

71.

Williams

Rycroft-Malone

Burton

(2017) Bringing critical realism to nursing practice: Roy Bhaskar’s contribution. Nursing Philosophy 18(2): e12130.

72.

Wilson

Janes

Williams

(2022) Identity, positionality and reflexivity: Relevance and application to research paramedics. British Paramedic Journal 7(2): 43–49.

73.

Wong

Greenhalgh

Westhorp

, et al. (2012) Realist methods in medical education research: What are they and what can they contribute? Medical Education 46(1): 89–96.

74.

Wong

Westhorp

Manzano

, et al. (2016) RAMESES II reporting standards for realist evaluations. BMC Medicine 14: 1–18.

75.

Yardley

Kinston

Lefroy

, et al. (2020) ‘What do we do, doctor?’ Transitions of identity and responsibility: A narrative analysis. Advances in Health Sciences Education: Theory and Practice 25(4): 825–843.