Distributed leadership as a predictor of employee engagement, job satisfaction and turnover intention in UK nursing staff*

Research shows that low job satisfaction and poor employee engagement are among the most frequently reported causes of high turnover in the UK health care context (Alarcon & 2011; Collini et al., 2015; Fasbender et al., 2019). With a predicted shortage of 100,000 nursing staff in the UK by 2028/29 (The King's Fund, 2020), it is critical to retain staff by increasing their engagement and job satisfaction to ensure continual delivery of high- quality patient care (West, Bailey, & Williams, 2020). One promising solution that strives to address this is the implementation of distributed leadership through the Shared Governance (SG) framework in the UK health care context. Abstract Aim: To investigate how distributed leadership via the Shared Governance programme influences employee engagement, empowerment, job satisfaction and turnover intentions among direct care nursing staff in a large UK hospital. Background: Increasing turnover rates and shortages of health care staff in the UK has called for interventions to improve employee engagement and job satisfaction. Methods: 116 direct care nursing staff were sampled in a mixed- methods explanatory sequential design. A maximum variance sample of 15 participants were subse-quently


| BACKG ROU N D
Research shows that low job satisfaction and poor employee engagement are among the most frequently reported causes of high turnover in the UK health care context (Alarcon & Edwards, 2011;Collini et al., 2015;Fasbender et al., 2019). With a predicted shortage of 100,000 nursing staff in the UK by 2028/29 (The King's Fund, 2020), it is critical to retain staff by increasing their engagement and job satisfaction to ensure continual delivery of high-quality patient care (West, Bailey, & Williams, 2020). One promising solution that strives to address this is the implementation of distributed leadership through the Shared Governance (SG) framework in the UK health care context. Although researchers have yet to agree on the constituents of distributed leadership, this paper adopts Tashi's (2015) definition of distribution leadership as a shared decision-making framework or practice applied by various staff members across multiple organisational levels. This differs from other types of leadership where it emphasizes leadership as a practice that focuses on influence and agency through interpersonal interactions rather than formal roles, responsibilities and actions (Harris & DeFlaminis, 2016). Within the UK health care context, distributed leadership is demonstrated through SG, which is a form of structural empowerment where nurses are given greater autonomy and control in their practice to facilitate organisational change and improve patient outcomes (French-Bravo & Crow, 2015).
Through a formalized support structure, SG facilitates an inclusive, collaborative and shared decision-making process with the aim of driving innovative quality and service improvements that benefit staff and patients (NHS, 2021). Staff are voluntarily appointed to a council to make decisions that improve long-term professional, organisational and patient outcomes (Hess, 2020). Through SG, opportunities are given to staff to meet with the Chief Nurse to present pertinent issues and learnings concerning new ways of working clinically and operationally, and to advance nursing as a profession (NHS, 2021).
Although there is emerging evidence on the impact of SG, this evidence is primarily studied in the American private health care system. While researchers have attempted to examine how distributed leadership is implemented in a UK hospital (Geoghegan & Farrington, 1995;Jackson, 2000), there is still a paucity of research to understand how distributed leadership influences outcomes such as employee engagement, job satisfaction and turnover intentions in the UK. In this study, job satisfaction is defined as the positive affect and attitude one feels towards their job, which influences personal fulfilment, sense of achievement and opportunities for recognition and promotion (Armstrong, 2006;Kaliski, 2007). Employee engagement is the cognitive, emotional and behavioural aspects of an individual who are directed towards desirable organisational outcomes (Shuck & Wollard, 2010). Turnover intention is defined as an employee's intentions to stay or leave the organisation in which they are employed (Bothma & Roodt, 2013).
To better understand how distributed leadership affects these employee outcomes, this paper reports findings from a study conducted in a large UK teaching hospital that implemented a SG programme that has applied distributed leadership since 2012. Specifically, this study explores how distributed leadership through the SG programme has influenced employee engagement, job satisfaction and turnover intention among staff in the UK health care system.

| Design and procedures
A mixed-methods explanatory sequential design (Creswell & Plano Clark, 2017) comprising a survey and semi-structured interviews was used to address the research question. A sequential approach of quantitative followed by qualitative methods (Morse, 1991(Morse, , 2003 was used due to the deductive nature of the research question. This allowed the a priori hypotheses of the effect of distributed leadership to be tested first through quantitative survey data and then complemented by qualitative interviews to deepen insights and explain the influence of distributed leadership on employee outcomes (Creswell & Plano Clark, 2017;Gutterman et al., 2015).
Participants were registered and non-registered direct care nursing staff 1 (DCNS) from an NHS Teaching Hospital Trust who demonstrated distributed leadership agency either formally via the SG programme or informally in their current roles. They were recruited via email, which contained a link to an online questionnaire, which took 30 to 45 min to complete. At the end of the questionnaire, participants who were currently involved in the SG programme were invited on a voluntary basis to participate in a follow-up faceto-face interview. Purposive sampling was used to select participants for the interview to achieve a maximum variance across demographics, roles and experience within SG. Prior to the interviews, an interview guide was designed to explore themes relating to the impact distributed leadership had on participants via SG while allowing other relevant themes to emerge during the interview. The length of the interviews ranged between 20 and 40 min, and they were conducted on hospital premises during working hours. All interviews were audio-recorded and transcribed verbatim.

| Participants
Due to the limited time to complete the study, a convenience sample of 116 DCNS completed the questionnaire. To achieve maximum variance based on the sample's gender, age, ethnicity, highest education level, job designation and length of time participating in the SG programme, 15 participants were recruited for the interview. Recruitment was stopped when data saturation was reached.
Participants who were interviewed were involved in the SG programme between 1 and 84 months (M = 19.60; SD = 21.71).

| Measures
The Distributed Leadership Agency (DLA) is a 7-item scale assessing involvement in leadership tasks (Jønsson et al., 2016;Unterrainer et al., 2017) comprising three dimensions related to change, tasks and relations (Yukl et al., 2002). Responses are measured on a 5point scale ranging from 1 = not at all to 5 = very much. The DLA was validated in a hospital setting and was shown to have high reliability, with Cronbach's α ranging from 0.91 to 0.93 (Unterrainer et al., 2017). 1 Direct care nursing staff are registered nurses who are responsible for assessing, planning, implementing and evaluating care of patients (e.g. treatments, patient education and administration of medicines). This also includes non-registered nursing auxiliaries that work alongside registered nurses to deliver patient care.
The Utrecht Work Engagement Scale (UWES) is a 17-item scale assessing levels of engagement at work (Schaufeli & Bakker, 2004) and comprising three subscales: vigour, dedication and absorption. Responses were scored on a 7-point scale ranging from 0 = never to 6 = always. The UWES was shown to have adequate reliability and good internal consistency of Cronbach's α between 0.80 and 0.90 (Schaufeli & Bakker, 2004).

The Minnesota Satisfaction Questionnaire-Short Form (MSQ-SF)
is a 20-item scale measuring job satisfaction (Weiss et al., 1967) and comprising three subscales: intrinsic satisfaction, extrinsic satisfaction and general satisfaction. Responses were recorded on a 5-point scale ranging from 1 = very dissatisfied to 5 = very satisfied.
The MSQ-SF was shown to have good reliability and discriminant validity between intrinsic and extrinsic job satisfaction in relation to other relevant variables in the MSQ (Hirschfeld, 2000).
The Turnover Intention Scale (TIS-6) is a 6-item scale assessing behavioural intention to leave the organisation (Bothma & Roodt, 2013;Roodt, 2004). Responses were measured on a 5-point Likert scale with higher scores reflecting a greater intention to leave. The TIS-6 was shown to have good criterion-predictive validity, differential validity and reliability of Cronbach's α = 0.80 (Bothma & Roodt, 2013).

| Analysis
Hierarchical multiple regression was used to test whether the addition of DLA (Step 2) obtained from a submaximal test improved the prediction of employee engagement, job satisfaction and turnover intentions over and above demographic variables (i.e. sex, age, ethnicity and highest education level) while controlling the effects of covariates (i.e. supervisory role, length of time working in health care and length of time involved in the SG programme) (Step 1).
Thematic analysis was used to analyse qualitative interview data. Braun and Clarke (2006)'s 6-phase approach was followed, comprising: (1) familiarization with data; (2) generation of initial codes; (3) search for themes; (4) review of themes; (5) defining and naming of themes; and (6) producing the report. Both deductive and inductive approaches were used during coding and analysis of data to ensure identified themes matched existing literature while allowing the emergence of new themes. Data collection stopped when data saturation was reached.

| Distributed leadership and job satisfaction
As indicated in

| Distributed leadership and turnover intention
As indicated in Table 2, in Step 1, the combined covariates did not predict turnover intention. The full model predicting turnover intention ( Step 2) was statistically significant (R 2 = 0.231, F(10,105) = 3.149, p = .001; adjusted R 2 = 0.157), and the addition of DLA led to a statistically significant increase in R 2 of 0.081 (F(1,105) = 11.077, p = .001). Table 3 summarizes the demographics of participants interviewed.

| Qualitative results
The themes identified in the data are described below and illustrated by original quotations.

| Engagement at work and in the profession
Participants reported that distributed leadership enhanced their engagement at work as it gave them opportunities to represent and share their peers' and teams' perspectives to influence departmental and organisational issues.

Turnover intention
Step 1 Step 2 Step 1 Step 2 Step 1 Step 2   Participants also expressed appreciation in being given opportunities to make a difference through their work, which led to a sense of personal meaning and fulfilment that increased their job satisfaction.

| Sense of belonging and commitment to the organisation
Through the interactions that distributed leadership facilitated, participants felt more connected to the wider organisation as it provided them opportunities to network and work with staff from other departments and job roles.

| Challenges associated with SG
While participants were generally more engaged, empowered, satisfied and committed to the organisation, a few challenges relating to the implementation of SG appeared to influence distributed leadership outcomes.
First, participants who were new to SG verbalized inadequate onboarding information that resulted in a lack of common understanding and communication regarding the purpose, structure and processes of SG. This appeared to affect their motivation and ability in executing ideas that were beneficial to patients, staff and the organisation.

'We don't know what we don't know. So, I don't know what we should be asking…communication could be better'. (P11)
While participants generally enjoyed their involvement in SG, a few of them expressed challenges in terms of conflicting priorities and time pressures in managing their formal job roles and SG responsibilities, which influenced their job satisfaction and mental well-being.
'You're given a small amount of time in your rota to do it.

But people expect you to be answering emails and chasing things up…it becomes an added responsibility or stress'. (P6)
Although there were structures and processes in place to support SG efforts, a few participants commented that the slow progression and laborious efforts involved in executing SG projects affected their motivation and commitment in completing projects expediently to maximize outcomes.

| D ISCUSS I ON
Quantitative results suggest that distributed leadership through SG significantly increased employee engagement and job satisfaction and significantly decreased staff's turnover intentions. Qualitative findings were consistent with quantitative results, presenting a unique perspective from the NHS Trust on how SG increased work and professional engagement, and empowered staff by giving them a voice to share their ideas and confidence to implement these ideas to improve patient, staff and organisational outcomes.
Findings from this study showed that being involved in distributed leadership tasks significantly increased employee engagement by 10.7% above that explained by demographic variables such as time in the SG programme. A systematic review by Beirne (2017) supports this, highlighting that a key feature of distributed leadership is in effective employee engagement that contributes to valu- As engagement is critical in addressing the current and projected shortage of nurses (West, Bailey, & Williams, 2020), it is recommended that hospitals incorporate processes and systems that encourage staff's active involvement in sharing ideas that can bring positive impact to patients, staff and the organisation (Hussain et al., 2018). It is also imperative that a psychologically safe environment is created for staff to share innovative ideas that are validated, and to provide opportunities and resources for them to execute their ideas to increase engagement and improve professional practice (Edmondson et al., 2016).
Findings from this study showed that SG empowered participants in making positive changes in the organisation. Using Kanter's (1993) theory of structural empowerment, Moore and Hutchison (2007) explain that distributed leadership increases participants' access to information and resources that aid their job, provides relevant support, widens their social network and gives them opportunities for professional development and growth. Because participants have increased access to knowledge, this naturally increased their confidence, motivation and self-direction towards developing themselves professionally (Bradbury-Jones et al., 2010;Friend & Sieloff, 2018).
As access to information, support and resources are important contributors to structural empowerment (Spencer & McLaren, 2017), it is essential that SG provides the relevant information and support required to empower staff in executing their ideas and projects. To address challenges of a lack of common understanding and communication regarding the purpose, structure and processes of SG, it is recommended that SG programmes implement an onboarding process to provide staff with relevant information about the programme, provide direction on where to get resources and support, and help facilitate the building of connections beyond their appointed SG council, which will set staff up for early success (Carucci, 2018).
Moreover, as high-quality leader-member exchanges have been found to increase engagement and empowerment at work (Aggarwal et al., 2020), SG council meetings should facilitate meaningful conversations that validate staff's contributions, showcase ways in which their contributions have made a difference and increase confidence in their capabilities to deliver beneficial patient and organisational outcomes.
As the UK health care staff consistently struggle with stress due to heavy workloads and feelings of being devalued (Senek et al., 2020), this study's findings are pertinent in addressing issues of disempowerment in the workplace. To address the increasing rates of burnout within the UK health care sector, it is recommended that organisations be transparent in their processes, increase engagement with their staff, increase management support and share information that will be helpful in facilitating staff work and enhance empowerment at the workplace. Where possible, organisations should also invest efforts in developing staff professionally and provide sufficient management support to create a positive work environment that enhances staff empowerment, which would further increase organisational performance (Markos & Sridevi, 2010).
To ensure that SG processes do not have a countereffect in slowing down the execution of ideas resulting in more work and stress, it is imperative that SG processes be reviewed regularly, streamlined and communicated to enhance the efficiency of practice (Bohman et al., 2017).
Additionally, results from this study showed that distributed leadership significantly increased job satisfaction by 12.6% above demographic factors and that staff empowerment also contributed to the increase in job satisfaction. Participants' positive appraisal of their work was found to increase job satisfaction, which was congruent with Bagozzi's (1992) attitude-intention-behaviour model, which explains that behaviour is a result of an individual's cognitive appraisal of various work domains leading to a positive emotional response to work (Zeinabadi, 2010). Moreover, literature asserts the integration of employee engagement, staff empowerment and job satisfaction such that when staff feel empowered and are engaged in dialogue to share their ideas for improvement, they often exhibit more creativity in problem-solving, which is beneficial in transforming organisations (Zhang & Bartol, 2010).
As job satisfaction has been shown to have an important role in reducing turnover (Fasbender et al., 2019), hospitals that implement SG should plan efforts around celebrating the successes of staff's contributions (Nursing Times, 2016) as it increases job satisfaction and provides staff with a sense of meaning and achievement, which contribute to behaviours that benefit patients, staff and the organisation (De Clercq et al., 2019).
As presented in the results, SG was found to significantly reduce turnover intention by 8.1% above demographic factors.
Literature has long proven job satisfaction to be a significant predictor of turnover intention in the health care sector (Jones, Warren, & Davies, 2015). Although qualitative findings did not explicitly highlight this relationship, participants' reports of feeling a sense of belonging and being valued as a staff member have been shown to increase organisational commitment (Dávila & García, 2012). Han et al., (2010) explain that because distributed leadership facilitates shared decision-making, it increases participants' psychological ownership in sharing their knowledge with their colleagues and fosters a strong sense of belonging, which increases organisational commitment. Moreover, findings suggest that organisational commitment and loyalty increased when participants got to know their colleagues better. Lee and Kim (2011) argue that because organisational attitudes and behaviours are socially constructed, an expansion of one's social network can inadvertently increase organisational commitment.
In view that an increase in job satisfaction can reduce turnover intentions, it is advisable for organisations to be mindful in engaging staff on issues that motivate and increase their satisfaction at work through interventions such as SG. Actions should also be taken to reduce conflicting priorities and time pressures that staff may experience due to their involvement in SG as it can lead to counterproductive outcomes of decreased well-being and burnout (Van de Heijden, Mahoney, & Xu, 2019). As SG is a voluntary appointment, clear processes should be outlined to allow staff to opt out and hand over their responsibilities if required (Johnson et al., 2016). Regular reviews should also be conducted to assess staff's desire to continue in SG to maximize effectiveness and delivery of positive outcomes (Demirkiran et al., 2016).

| Limitations
While this study provides insight on how distributed leadership influences employee outcomes through SG, several limitations are observed. First, while the measures used in this study were reliable and valid, there was a lack of studies validating these measures in the UK context, which may have resulted in the inflation of statistical models.
Second, although the DLA scale is one of the most current and reliable tools in measuring a participant's involvement in leadership tasks, it was limited in its scalability in determining whether participants had low or high DLA. This possibly impeded the depth of analysis and interpretation of results in relation to employee engagement, job satisfaction and turnover intention.
While efforts were made to include participants with a range of experience in SG, majority of participants had less than 24 months' experience in the programme, which may result in potential biases in the results.

| CON CLUS ION
In summary, this study can be considered one of the first in providing valuable insights on how distributed leadership in the form of SG influences health care staff, specifically in the domains of employee engagement, empowerment, job satisfaction and turnover intentions in a UK context. As employee engagement, staff empowerment, job satisfaction and turnover intentions are concepts that are intricately connected, this study provides hope in utilizing distributed leadership to address the struggles the NHS face in meeting the growing demands of recruiting and retaining quality DCNS to enhance care delivery.
Although the UK is in its infancy stage in rolling out SG across health care systems and more research can be conducted to explore its effectiveness, efforts such as the development of a national SG council provides hope in implementing distributed leadership as a solution to enhance employee outcomes and improve patient care within the nursing profession (NHS England, 2019). To further encourage the implementation of SG across the wider UK health care system, knowledge management processes can be set up to facilitate sharing of best practices, success stories and learning outcomes during health care conventions and events to increase buy-in (Intezari et al., 2017). Rotation programmes can also be leveraged for more experienced nurses to share their knowledge and experiences and help in the initial setting up of SG structures and processes for health care organisations that indicate interest in implementing distributed leadership (Thompson et al., 2004).
To further enhance the quality and effectiveness of data for similar studies, researchers may consider collecting organisational data such as organisational health survey results and actual turnover rates and triangulating these data during the analysis phase. With the triangulation of data, practitioners and legislators will be better equipped in formulating practical interventions for implementation of distributed leadership at the workplace.

| IMPLI C ATI ON S FOR N UR S ING MANAG EMENT
Findings from this paper suggest that distributed leadership in the form of SG is a promising solution to address high turnover and staff shortages in the UK and beyond. In addition to encouraging the adoption of distributed leadership in hospitals, nurse managers can provide staff with equal opportunities in contributing ideas and solutions that improve patient care and enhance the effectiveness of hospital policies, processes and practices. Nurse managers can also use distributed leadership as an opportunity to enhance staff development by empowering them to lead new and innovative approaches to advance health care and staff well-being. In advocating distributed leadership, staff will feel more empowered to champion and execute ground-up initiatives, possibly increasing their engagement, job satisfaction and commitment to the organisation. Through these, issues of high turnover, staff shortages and job retention can be better addressed.

ACK N OWLED G EM ENTS
We would like to thank the University of Nottingham and the Nottingham University Hospital NHS Trust for supporting this research. We would also like to thank our families for their unconditional support, without whom we would not have been able to complete this paper.

Conflict of Interest
All authors would like to declare that there was no conflict of interest pertaining to this research study.

Ethical Approval
The Research Ethics Committee of the University of Nottingham,

Faculty of Medicine & Health Sciences, approved this study with
Reference Number: 268-1903.