South WestLeadership Academy

Leadership & Leadership Development in the NHS:

A Short Review

www.southwestleadership.nhs.uk

Authors

Dr Philip Miller Dr Lucy Sitton-Kent

2 3

On behalf of the ten regional NHS Leadership Academies and the NHS Leadership Academy, South West Leadership Academy has commissioned this review into leadership and leadership development from the East Midlands Academic Health Science Network.

NHS South West Leadership Academy is committed to developing outstanding leadership in health in a consistent, systematic manner to ensure there is sufficient scale and pace to deliver the leadership required for the transition and to lead the new system within a tough financial regime requiring high levels of innovation, engagement and devolution. We have a long and proud tradition of providing unique and bespoke leadership development opportunities to the South West, which can’t be found anywhere else in the country.

Christina Quinn Director NHS South West Leadership Academy

Foreword ........................................................................................................................ 2

Executive summary ................................................................................................ 4

Findings ........................................................................................................................... 5

Introduction .............................................................................................................. 11

Method ......................................................................................................................... 13

Question One .......................................................................................................... 14

Question Two .......................................................................................................... 25

Question Three ...................................................................................................... 33

Summary of recommendations from the literature ................ 49

Conclusion .................................................................................................................. 50

References .................................................................................................................. 51

About the authors ............................................................................................... 61

Foreword Contents

4 5

The NHS is heading into uncharted waters in terms of increasing demand for more responsive services for an ageing population within a climate of recent, current and likely future financial constraints. The need for integrated services means that the NHS must:

n work with, not merely alongside other services

n maintain wellbeing, not just intervene in illness

n work for the public, not just for current patients

This is a transformational agenda, which will need strong, consistent and resilient leadership.

As Ham and colleagues note the NHS must urgently decide:

“ …how to combine work on leadership development and quality improvement as part of a coherent programme of development and support.”

(Ham et al. 2016:13).

Historically, leadership development in the NHS has been aimed at relatively small groups of senior staff within organisations with the intention that they be equipped to drive quality improvements from the top. In contrast, more recent work on leadership has increasingly drawn on evidence that distributed and collective leadership models are also required to deliver safe and high-value care; that is, leadership is also needed much closer to where care is actually delivered.

This new paradigm simply means that at whatever organisational level quality improvement is needed, is also the place to identify, develop and support leadership. The inherent implication in this is that a broader scope and increased demand for leadership development will be created with a move away from simply investing in existing figureheads, towards a strategy to generate enthusiasm, competence and responsibility for achieving excellence throughout the NHS.

This short review examines recent evidence for the most effective leadership development strategies to support clinical quality improvement, the kinds of indicators which can be used to assess the impact of leadership development programmes and to whom and when leadership development might need to be delivered in the future.

Findings: n Attitudes which value and foster empowerment and supportive nurturing cultures also enable

safety and quality improvements

n Developing leaders’ understanding of followership could improve the effectiveness of essential subordinate-superior working relationships

n Positive impacts on patient level outcomes such as safety and compassionate, responsive care tend to emerge through “cultural” processes requiring leadership qualities at senior management and board level

n Effective leadership is associated with improvements in intermediate outcomes such as staff wellbeing, engagement and turnover intention and these in turn have been shown to lead to improvements in patient outcomes and experiences

n Some direct improvements in clinical outcomes (e.g. mortality or readmission rates) from leadership interventions have been demonstrated but these findings are contested on methodological and theoretical grounds in terms of cause and effect

n The most direct impacts of leadership on quality of care are seen where leadership development is directly focussed on increasing capability in quality improvement and evidence-based practice techniques at all levels of an organisation

n Shared or distributed leadership is essential to achieve improvement of healthcare delivery processes due to the complexity patient journeys and the distribution of technical and professional knowledge.

n Many different NHS staff need to operate as leaders from time to time, to both drive improvement and prevent lapses in the quality and safety of care

n Increasing the skills of leaders in developing, mentoring and coaching others, can magnify the impact of leadership development down and across an organisation or department and are essential to achieve service-wide impacts

n Succession planning is important to maintain effective leadership over time and thus sustain progress in quality improvement

n Leader-member exchange theories are important but underused in healthcare leadership research - such theories increase understanding that staff relationships and “emotional intelligence” rather than the leadership qualities of individual staff alone, may largely account for effective working and quality improvements

n Leaders who employ appropriate styles in differing situations achieve better performance and increase their sphere of influence in nursing and across “professional bureaucracies”

n Skills in implementing evidence-based healthcare and the judicious use of performance indicators can help to align objectives and increase engagement of diverse staff groups to raise quality and improve safety

n Consistency of leadership and perceived commitment are key to driving quality improvement

n The clinical leadership development needs of staff in isolated community settings such as primary care, community teams and nursing homes, have been neglected despite the increasing movement of care away from acute settings and into community settings

Executive summary What leadership development strategies best support clinical quality improvements?

6 7

n Realist interpretations of the impact of leadership development on more subordinate staff groups lacking formal authority, suggests that focussing on maximising the influence of followers in “human factors” terms is essential for them to act effectively in support of improvement initiatives

n There is little or no research evidence of the impact of leadership development to support patient and public involvement - such needs/applications are not considered within the current body of leadership research

n Some aspects of organisational performance such as error reporting are clearly associated with open and supportive leadership cultures and encouragement of organisational learning

Recommendations from the literature:Core syllabus for leadership in health should include:

n Self-awareness and understanding self and others’ personal traits and their effects on forming and maintaining effective relationships and exerting influence effectively

n Working alongside other professional and managerial leaders to reduce “disconnects”, “siloed” thinking and negative stereotypes between professional groups

n Understanding that shared goals and perspectives are essential to achieve improvement

n Theoretical and practical understanding of how teams function and how this can be driven by emerging insights from research into followership, and shared and distributed leadership

n Practical applications of research into transactional styles of leadership to increase understanding of what type of followers they have, need or could create

n Understanding the importance of their role in developing staff and in particular potential successors, to sustain improvement gains

n Understanding that variable and responsive leadership styles and actions are required to build and maintain relationships with colleagues across boundaries and over time

n Skills in creating and using high-quality information to measure and sustain improvements by allowing “the data to talk” to staff with both clinical and managerial perspectives

n Understanding of the importance of patient perspectives and involvement as vital for successful transformational change

n Understanding the role of leadership across systems and care pathways involving multi-professional groups

n Leadership development should include when and how leadership shifts from employee to employee or level to level, both formally by delegation and informally via contextualised expertise

n Subordinate staff are often already well disposed towards change with which they can identify – leaders should be helped to harness such attitudes to drive quality improvements under collective or distributed leadership models

Findings: n There is little high-quality evidence that patient care outcomes or other high level indicators can be

used to measure the effect of leadership development directly

n Much of the available research relies on self-reported or observed leadership behaviours generally without control groups

n Intermediate processes and outcomes such as human resource metrics e.g. turnover intention are more likely to give useful information

n Validated measures of intermediate outcomes have been developed – some examples include “shared governance” assessment tools, 360o feedback or staff-level metrics such as identification with organisational aims

n Fewer, more consistent and better validated instruments should be used to assess the leadership capability of individuals or within specific departments or organisations

n Qualitative data derived from research, internal audits and inspections can also be used to assess the quality of leadership but should be carefully analysed

n Integration of as many sources of information as possible achieves a balanced view of the leadership “culture” within an organisation

n There is good evidence that measuring improvements in terms of intermediate outcomes such as safety culture, or reduced turnover could link leadership development to patient outcome improvements via cultural mechanisms and theory

n Longitudinal as opposed to cross sectional evaluations produce stronger evidence of cause and effect whether on individual or institutional progress

n Empirical evidence of the effectiveness of leadership training is largely limited to self-reported outcomes The quality of mentor-mentee relationships appears to be important and should be monitored

n Direct observational evidence and feedback could indicate where barriers exist to enacting leadership in the workplace despite theoretical understanding of leadership from development programmes

n Successful distributed leadership may still require policies to be sanctioned by senior management to create empowerment where centralised and hierarchical leadership models persist

n Well-designed comparative evaluations with suitable controls are needed to demonstrate the effects of specific leadership development interventions to underpin assumptions regarding cost-effectiveness

n Increasing the adoption of a limited number of validated indicators based on strong theoretical foundations e.g. Leader-Member Exchange theories or shared governance models is vital to demonstrate the direct impact of leadership development programmes

Which indicators can be used to assess the impact of Leadership development programmes?

8 9

Recommendations from the literature:n Leadership training must emphasise the importance of active feedback seeking behaviour - “how’s

my leadership?” – alongside more common self-reported indicators

n Effective leaders should share an understanding with their staff of both what is working, what is not and why - their success in doing so should be measured e.g. CQC key metric “well led?”

n Leadership development initiatives should be evaluated using objective intermediate outcomes such a “shared governance” metrics, measures of organisational culture, measures of staff well-being and longitudinal process measures such as incident or safety reporting

n Some longitudinal patient-centred indicators such as mortality can be difficult to interpret but may be useful for indicating effective distributed leadership over time

n Intermediate indicators such as numbers of reported incidents or drug errors may initially rise in well led organisations with “open” and active safety cultures but over time the number of similar or repeated incidents should then fall as effective leadership leads to learning

n Many existing human resource indicators can shed light on the effectiveness of leadership for example skill mix or turnover intention may warn of potential effects on patient experience even if direct improvements are harder to demonstrate

n There is evidence that leadership development should precede or coincide with significant career transitions into roles with leadership demands e.g. at graduation or on progression to junior consultant, ward manager or team leader

n Evidence from personal Development Reviews and other internal data collection could provide evidence of efficacy of development and to target the development needs of staff groups for succession planning

Findings: n Recent policy and research has consistently identified shared or distributed leadership models as best

able to deliver quality and safety improvement in high-complexity industries such as health care

n Artificial divisions between clinicians and managers and data–driven performance management approaches which foster such divisions should be de-emphasised and scaled back

n Many clinical staff already have responsibilities for “management” functions and therefore have scope for significant influence on clinical care via direct and indirect mechanisms

n A wider group of potential leaders must be identified and developed to align with organisational objectives and ensure improvements are delivered by “hands-on” staff

n Interprofessional leadership training is still rare despite the acknowledged need for greater collaboration between clinical and non-clinical leaders

n Inter-professional training best supports collective leadership and can help to overcome negative attitudes and preconceptions resulting from latent structural divisions

n Disconnects, misunderstandings and conflicts at senior staff level can profoundly affect organisational culture and therefore how subordinates interact and collaborate

n Patient leadership development initiatives are rare but demand is likely to grow

n Indirect evidence suggests that patients and their families may have unique leadership development needs

n Evidence from followership research has highlighted the importance of shared goals, genuine and sustained engagement and the importance of common, high-level goals

n The “patient”, as a high-level concept, provides a simple way to understand how alignment can be achieved across professional boundaries and negotiation between competing imperatives

n Gender can be of great importance in the relative success of mentoring and coaching - for example there is evidence that male mentors underestimate or misinterpret the developmental needs of female mentees. Leadership development should include the importance of gender in enabling all staff to increase their own effectiveness

n The findings from followership research suggest that altering leader-member relationships should be undertaken carefully as unintended consequences for performance and wellbeing can occur at different levels of seniority

n There are large staff groups in less regulated environments, for example in nursing home and community settings, who currently receive little or no leadership training or support despite having a key role in delivering quality care

n More research is needed to understand the optimal staging of leadership development to best support succession planning and sustained improvement

n Key professional transitions e.g. from junior doctor to consultant role are opportunities for leadership development and early access should be a primary goal

Can leadership development be better targeted? Is there evidence of impact for single or multi-professional groups, in certain settings or at specific levels of seniority?

10 11

Recommendations from the literature:n Leadership development for senior staff must reflect inter-disciplinary and multi-disciplinary

perspectives and how effective alignment can best be achieved amongst followers

n Shared leadership and team training relevant to seniority and professional group should be prioritised but is not well understood and should be rigorously evaluated

n Leadership development should focus on showing people how to create “direction, alignment and commitment” despite superficially competing goals

n Staff engagement at all levels should be a key aim of leadership development but may require targeted support for less empowered groups

n Project and service management activities requiring team leadership should explicitly draw upon distributed or collective leadership theory to understand how quality improvement is realised

n Specific leadership development interventions should be used to increase the effectiveness and supply of leaders drawn from patient and family cohorts to ensure patient-centred improvement across the NHS

n Patient and public voices should be direct sources of inspiration for leaders and should take centre-stage in leadership development aimed at increasing the value of care

n Understanding of the causes and effects of “good” (committed and engaged) and “bad” (reluctant and obstructive) followership is required in complex and highly technical organisations such as those delivering healthcare

n Leader-member exchange theory should be used to create followership development opportunities

n Followers should be helped to understand their responsibilities and influence in achieving successful working partnerships rather than exclusively attributing responsibility for failure to individual leaders

n Leadership development should aim for improved understanding of team functioning, effective delegation and communication and the importance of “human factors” in delivering quality “hands-on” care

n Mixed professional leadership development programmes should cater for the differing needs of professional groups by acknowledging the effects of existing hierarchies, barriers, degree of self-efficacy and realistic expectations

n Patient-centeredness should be a central component in cultivating authenticity in NHS leadership

n Development programme leads should recognise and address findings from research regarding gender as directly affecting leaders’ ability to successfully develop their own subordinates and potential successors

n Relatively isolated but key staff such as care home managers, community team leaders and general practitioners should receive appropriate support to exercise leadership

n Standards for leadership should be incorporated into commissioning processes and evaluations as they already are within CQC inspection criteria

It has long been recognised that the NHS, requires “great leadership” to navigate these conflicting pressures successfully. Popular opinion, political rhetoric and media commentary all claim to distinguish NHS leadership from mere management. Peter Drucker, an early management theory guru, wrote that:

“ Management is doing things right; leadership is doing the right things. Management is efficiency in climbing the ladder of success; leadership is about determining whether the ladder is leaning against the right wall”.

Peter Drucker in (Frankel 2008:1)

Frankel uses Drucker’s quote to suggest that management is about tasks and performance whereas leadership demands judgement or “philosophy” (Frankel, 2008). It is therefore tempting to assume that the future of the NHS requires leadership of the individual and philosophical variety to enable it to focus limited resources on doing the “right things” and scaling the “right walls”. But it is increasingly clear that to enact radical change a majority of NHS staff must act as leaders by shaping and enacting transformative change whilst ensuring that high standards are maintained and resources used wisely.

Recent publications and research have begun to argue that both centralising “leadership” and dividing it from “management” could be actively harmful to the NHS and its patients. The task of the NHS is to deliver healthcare which leads to increased wellbeing and it is hard to maintain that any decision taken at any level has no impact on quality however small. Therefore the role of leader could and should be open to anyone who wants better care for their patients. Similarly the evaluation and “management” of care processes and outputs should be everyone’s business. Greater acceptance of these goals has led to calls for “distributed”, “shared” or “collective” leadership and a renewed emphasis on the importance of followers to ensuring delivery.

The NHS context is a highly technical, integrated and complex service with a variety of sources of knowledge, power and authority. Many staff combine clinical and managerial roles within their day to day work. In addition the NHS is increasingly guided by a combination of research evidence and evaluations of services using valid indicators. Due to the complexity of the context the NHS of the future cannot rely on individuals or small group of leaders to deliver changes at scale and pace. Rather leadership must be recast as effective followership at many levels and within many disciplines, for sustainable service improvements to be achieved.

Introduction

12 13

The future task of leadership development is therefore to enable leadership to emerge wherever and whenever needed to meet these new expectations. This new approach recognises that leaders need to be all those variously labelled as managers, clinicians, technicians as well as executives and directors. The implication of this new thinking is that the majority of NHS staff should understand the influence they have on those around them and exercise it wisely. Recent research high-lights the importance that the behaviour and attitudes of followers have on their leaders. This is a profound point: if an individual can (or could) change the behaviour or attitudes of other NHS employees for the better, then that individual is a leader. However, targeting leadership development at “everyone” is unsupportable. But where individuals are given access to leadership development opportunities they should be designed to ensure a thorough understanding of newly important concepts such as collective leadership, talent management, succession planning and followership, as essential ingredients in a transformed and integrated NHS.

The following short report assesses recent academic evidence about effective leadership development, its relationship to high value care and where it might best be targeted. The report is structured around three related questions agreed with the commissioner:

1. What leadership development strategies best support clinical quality improvements?

2. Which indicators can be used to assess the impact of Leadership development programmes?

3. Can leadership development be better targeted? Is there evidence of impact for single or multi-professional groups, in certain settings or at specific levels of seniority?

A short review of recent literature was undertaken in addition to a focussed reading of recent academic reports and policy documents. The executive summary describes our findings and collates recommendations that emerge from this evidence base. The main document mirrors this structure and summarises some of the live debates about the future of leadership development for improvement in the NHS. The review process employs principles and techniques drawn from Pawson (Pawson et al. 2005) which are useful when investigating a broadly defined and diverse topic or one supported by a diverse literature base. Sources were identified in a process moving from initial “keyword” proposals, through iterative searches of academic and grey sources to a synthesis of material designed to give an accessible overview. The editorial, literature reviewing and commissioner feedback processes were conducted in tandem to ensure the eventual review met the requirements of the brief whilst achieving a satisfactory degree of rigour, accessibility and timeliness.

The key terms and working definitions for this review are as follows:

Clinical Leaders = staff with explicit leadership or management roles whose primary qualifications are clinical and whose early career was within a clinical profession: medicine; nursing; AHP etc.

Clinical Leadership = operational, management or purely leadership activities carried out with a focus on improving clinical care and outcomes

Leadership developmental intervention = training or other specified programme or educational approach aimed at increasing the ability of student to perform clinical leadership functions and behaviours

Leader Development = focused on specific individuals and primarily designed to increase their proficiency at various components aimed at influencing subordinates

Leadership Development = programmes or initiatives of any kind or content designed to improve the effectiveness of leadership within a defined organisation or across a system

Follower = anyone answerable to someone else for part or all of their role

Followership = the abilities and processes of influencing and being influenced by leaders

The following sections provide the background to the findings and recommendation outlined in the executive summary.

Method

14 15

The ultimate aim of leadership development is to ensure that the highest quality of care becomes the sustainable norm across the organisation or department or staff group over which the leader or leaders have influence. There are several abilities identified in the literature as central to enabling strong and consistent leadership. These can be divided into those that are within the organisation’s control and those that are individual characteristics. A variety of programmes, individual components, activities and learning strategies have been incorporated into the various leadership development initiatives focusing on these abilities. The following section examines recent evidence regarding the effectiveness of some of these abilities and the degree to which they have been shown to lead to improvements in clinical care and service performance. This area of leadership research is undermined by significant methodological flaws meaning that many associations between leadership, performance and quality remain contentious (West et al. 2015). Both evidence of direct impacts on patient care and intermediate outcomes thought to be important in achieving such goals are included where available.

OrganisationalA supportive environment

One method of preparing leaders for such quality improvement processes is to use “Action Learning Sets” (ALS) to either simulate or more commonly actually undertake, evaluation and quality improvement projects. These are conducted in a supportive and learning oriented environment which is more or less separated from the normal working environment or where external facilitators and mentors enter the participant’s workplace. ALS are generally defined as small groups of staff working through a change project governed by ground rules on behaviour with group direction sometimes facilitated by a more experienced and often external figure. Young et al. (2010) describe the use of ALS to develop “strategic” thinking skills in consultant nurses in two co-located Trusts in the NHS. Amongst the benefits identified were the actual improvements in care pathways and change initiatives considered along with increased self-confidence amongst participants. The latter eventually enabled them to influence initiatives at board, national and international levels and therefore leaders within and external to their organisation. In addition, participants felt themselves able to act as a “ready-made” support network for staff recently appointed to senior nurse posts thus supporting the emergence of future leaders within their organisations. This focus on talent management is an important element of the impact of programmes but by definition is hard to evaluate given the necessary timescales.

Solansky evaluated two elements of leadership training common to many programmes: 360o “leadership assessments” and mentoring (Solansky, 2010). The research involved 303 trainees and 41 programme mentors. Self-reported achievement and behaviour data significantly differed from the feedback from co-workers and subordinates for some trainees. Those leadership trainees who underestimated the quality of their leadership, by comparison with feedback from their co-workers, demonstrated greater improvements in actual performance and work achievements than those whose self-ratings were higher than co-workers. An effect of mentorship was also detected. The amount of time spent by mentors on coaching mentees was correlated with their improvements and skills improvement with similar effects for greater mentor-initiated personal contact. The authors recommend that supervision of the mentor-mentee relationship should form a part of all leadership development programmes so that this key relationship can be monitored and if necessary corrected.

Developmental Assignments are sometimes used to prioritise learning by delivering more theoretical content (West et al. 2015) although the authors stress the potentially demotivating effects of “dropping people into deep water” and no evidence of relative effectiveness is presented. Avolio reviewed the evidence on the effectiveness of leadership interventions in creating performance improvements including 200 “lab and field” (sic) studies (Avolio et al. 2009). The majority of studies involved a training or development intervention although others were based around role play, scenarios or vignettes or changes in leadership assignment (i.e. relabelling staff as leaders without other changes to their actual role). The vast majority of studies examined affective, cognitive or behavioural outcomes with only two examining organisational performance. There was some evidence that programmes could enhance performance at the organisational level and most programmes were cost-effective. It is not clear that this effect is independent or merely demonstrates that companies willing to invest in their staff are those that are better run overall.

Increasing engagement

The “engagement” of staff in both their own work and the wider aims of their organisations is now seen as a fundamental prerequisite for high quality, cost-effective care and therefore, a key objective for health service leaders (West and Dawson, 2012). Mulla and colleagues collected interview data from a number of senior managers and clinical staff in four separate services provided by three NHS acute hospital trusts (Mulla et al. 2014). They found that, despite wide acknowledgment of the need for clinical leadership, work to explicitly support clinical leaders and “champions” was limited in most areas. Instead the majority of the organisations studied expressed the belief that clinical leaders would “emerge” presumably as a result of their individual qualities and ambition. This attitude tended to be related to the “culture and priorities” of the different organisations and was not related to any specific work in the better Trusts to ensure that this happened in practice. Only one Trust reported any specific engagement initiatives which included the executive team meeting all newly appointed consultants during their induction to establish lasting communicative relationships.

Chapman attempted to measure the impact of leadership on the organisational “climate” for improvement in the context of two contrasting leadership theories; “transformational” and “distributed” (Chapman 2014). The study used qualitative and observational data from a variety of NHS organisations and senior staff. No direct measurement of patient outcomes or organisational performance was attempted but respondents expressed the belief that “good” leadership, defined as emotionally intelligent and creating staff engagement, was necessary to deliver high-quality care in a safe environment. Shipton and colleagues used a scale of leadership effectiveness, external hospital assessment results and a measure of “care quality climate”, to understand how they were interlinked in 86 NHS hospitals (Shipton et al. 2008).The research is unusual in examining patient complaints (number per 1000 patients and from inpatients, outpatients and EDs) as important indicators of quality, arguing that even in the absence of actual clinical negligence, they can often indicate poor communication. They suggest that positive effects of leadership on quality of care rest on the extent to which leaders succeed in influencing their staff to place the needs of patients first. The report finds that greater leadership effectiveness is associated with fewer complaints and higher care quality climate.

Question One: What leadership development strategies best support clinical quality improvements?

16 17

A further large study has examined the links between organisational cultures in NHS Trusts with outcomes related to patient safety and staff well-being (McKee et al. 2010). This analysis builds on existing theoretical links between quality of care and health care staff experiences as discussed briefly in other sections of this report, but focusses on the leadership of senior staff only. No direct associations were found between senior staff management style and organisational performance measures such as relative mortality (p.122). Senior leadership style was not associated with safety performance but signalling safety as a priority to the wider organisation was seen as “critical”, as was “dispersion” of leadership across the organisation (p.187). There was little evidence for an association between senior leadership and staff or patient outcomes such as injury but some support for a connection with incident reporting (p.192). There was strong support for a link between higher levels of support for clinical staff and lower staff-reported measure of stress, intention to leave and witnessing of errors and near-misses and higher job satisfaction (p.194). Overall the report suggests that leadership at different levels has measurable impacts only at that level. Resilience of a whole organisation to external “shocks” and delivery of patient level quality improvements rely on high-quality leadership at very senior and relatively junior levels respectively.

There is stronger evidence for collective leadership approaches in implementing successful improvements and creating a culture of quality improvement. It is unlikely that such cultures occur spontaneously and therefore a significant impact of effective leadership comes from enabling such cultures to emerge and be sustained rather than on any direct impact individual leaders have on complex organisational outputs or overall performance (West et al. 2014). Collective leadership is poorly defined and is often synonymous with staff engagement which for much of the leadership literature is an outcome and not itself defined as leadership (Ham, 2012).

Richardson and Storr (2010) undertook a review of the literature examining the potential for nursing leadership development programmes to improve patient safety. They found little high quality evidence to support the role of nurse leadership in improving safety directly. This was simply because the primary outcome of most of the studies was voluntarily reported incidents or medication errors. These are insensitive indicators, as it is estimated that no more than 15% of errors result in reports but are also highly influenced by the prevailing leadership and cultural practices themselves. The review did identify studies showing that the introduction of guidelines and checklists did enhance nurses’ ability to intervene and improve practices specifically when dealing with powerful colleagues. Rather than demonstrating nursing leadership it underscores its apparent absence in many contexts as demonstrated by Martin and Waring (2013). It may be that leadership development for less formally powerful groups such as nurses needs to enable them to use less formalised or overt levers such as appeals to research evidence or organisational norms, to influence others.

Some of the best quality evidence for the importance of leadership for increasing the quality of care comes from a large-scale multi-model study reported by Dixon-Woods et al. (2014). A variety of evidence is marshalled to demonstrate that effective leadership ensured that senior staff actively sought to understand “shop floor” performance, empower clinical teams and foster openness and accountability at every level of their organisation. They had such effects by setting clear “mission, direction and tone”, supporting and encouraging change and innovation and fixing system level problems. Barriers to improving quality were found, for example where leaders sought comfort in falsely reassuring indicators and performance failings as arising from individual front-line clinician behaviours.

The report summarises the most effective senior leadership strategies for developing such cultural strengths are:

n Continually reinforce an inspiring vision of the work of their organisations

n Promoting staff health and wellbeing

n Listening to staff and encourage them to be involved in decision making, problem solving and innovation at all levels

n Providing staff with helpful feedback on how they are doing and celebrate good performance

n Taking effective, supportive action to address system problems and other challenges when improvement is needed

n Developing and modelling excellent teamwork

n Making sure that staff feel safe, supported, respected and valued at work

(Dixon-Woods et al. 2014)

Staff retention and Skill-Mix

It is well recognised that staff retention and the resulting improved staffing levels and skill-mix have important implications for patient safety. A large well conducted study for example has demonstrated associations between intermediate human resource measures such as levels of self-reported bullying and Trust level measures of quality of patient care, mortality and performance outcomes (Powell et al. 2014). Outcomes included patient satisfaction, mortality and nosocomial infection rates. Some effects were stronger for some staff groups. For example nursing absenteeism and turnover was more strongly related to self-reported well-being and bullying than for other groups. Some directional effects were seen such as good staff experience causing lower absenteeism. There was evidence of a strong influence on improved staff experiences for variables associated with leadership such as staff reporting that they felt they could influence their organisation and that there was good communication with seniors. Little or no direct and positive causal links were found with mortality or infection rates. A positive association between good staff experiences and lower mortality was seen but there was no evidence that the former led to the latter despite sophisticated analysis designed to demonstrate this.

18 19

IndividualTechnical or clinical knowledge

Technical knowledge is widely thought of as necessary if not sufficient to exercising leadership for quality improvement. The degree to which expert knowledge is required appears to vary across contexts. For strictly clinical matters, professional membership and training is considered sufficient but in other contexts leadership must include responsibility for managerial knowledge and this presents a challenge if this is seen of lower importance.

In the case of IT implementation there is evidence that technical knowledge needs to be quite specific to enable leaders to drive improvement. Leadership is given prominence in frameworks regarding IT implementation because of the inherent problems and the degree to which IT affects multiple departments and staff groups (Ingebrigtsen et al. 2014 and Avgar et al. 2012). Leaders must ensure that all levels of an organisation align, from strategic planning through operational implementation to front-line responsiveness of the workforce, if IT projects are to perform as planned. Such projects require leadership across boundaries and beyond formal authority structures. “Soft” power and persuasion are associated with success more often than “command and control” approaches given the need for organisational learning. Leadership from those with specific IT skills, knowledge or enthusiasm for IT was associated with successful project implementation. Only one study reported on the effect of more generic leadership qualities. This found that leaders who were “self-confident, stable through adversity, steadfast, tough, and rich in initiative” were more likely to guide a successful IT projects to completion. Leadership training that emphasises delegation and gives the confidence for non-specialists to cede some authority would make leaders more effective in such situations.

Much of the IT literature is framed in terms of heroic and senior leadership paradigms and does not specifically address leadership training or more contemporary models of distributed leadership. The communication of clear goals and understanding of the vision and structures of IT governance, were associated with successful projects. It remains unclear the extent to which generic leadership capabilities are important or whether success stems mainly from superior IT-specific knowledge or experience. The ability to align IT implementation with existing patterns of professional working was noted as a positive skill and goal alignment is a recognised leadership characteristic. Understanding existing structures and workflows and managing change of and within them, are regarded as skills important to effective leadership more widely. Only one longitudinal study reported that some blame for the failure of a project lay with a lack of clear high-level leadership alone. These findings imply support for collective leadership models which are based on the shifting or sharing of leadership with topic expertise rather than relying on top-down authority.

Clinical excellence clearly requires clinical expertise and this understanding sits behind the repeated calls for more “clinical leadership” within the NHS. A leader’s ability to frame quality improvements in terms of acknowledged experience and by preference a clinical evidence base, is therefore key. Promoting this form of leadership recognises and legitimates the extent to which authority resides in or arises from bodies of academic and technical knowledge within healthcare. Leaders who can anchor improvement drives in clinical evidence could have significant influence. Reichenpfader et al. (2015) reviewed recent literature on the effect of leadership in evidence-based medicine (EBM)

but found little research supporting direct benefits on patient care. Some studies measured intermediate outcomes but often failed to give a theoretical explanation of the proposed end effect on either innovation or evidence-based practice (EBP). Over half of the articles examined were from nursing contexts and evidence regarding the effects of medical leaders was lacking. As with much previous research into leadership, there was a focus on the qualities of individual leaders at the expense of leadership processes or collective leadership. The extent to which EBP can be used to increase followership is therefore unclear but the need for leaders to be able to influence professional bureaucracy remains and the use of EBM is clearly appealing but is not sufficient to reliably create improvement on its own.

Staffileno and Carlson (2010) assessed the importance of EBP skills for developing effective leadership within nursing. They note that such skills are less developed amongst “direct care nurses” and yet the impact of such understanding could become significant as they move into positions of leadership. Nurses in leadership positions need to influence junior colleagues, peers and colleagues from other professional groups such as doctors. To effect change and improvement, skills in critiquing research, using data and rejecting “tradition-based” approaches to patient care could all enhance the impact of those in leadership positions. Flodgren conducted a review of how clinical leaders could influence professional practice and patient outcomes directly (Flodgren et al. 2011). The review found that a mixture of approaches had been tested using compliance with evidence-based guidelines as an outcome. Overall, “single opinion leader” interventions were found to increase compliance with best practice by 12% compared to an 18% improvement for multidisciplinary team approaches.

A recent randomised mixed-methods study has compared an EBP-focussed leadership development programme (Leadership and Organisational Change for Implementation; LOCI) to a more conventional programme delivered via webinar (Aarons et al. 2015). The programme focuses on relationship skills and achieving performance standards and is aimed at first-level leaders i.e. those directly responsible for leading care providers, in this case in mental health nursing. It included didactic teaching supplemented with coaching and organisational strategy development. Self-reports suggest that the LOCI group rated the feasibility, acceptability and utility of the training programme more highly than controls. In addition the supervised clinicians rated LOCI trained leaders more highly in their support for EBP than controls but no difference in “readiness” for it. The study results are weakened by the great dissimilarity in intensity of the two conditions (LOCI 6-month programme vs 2 1-hour webinars) and no intermediate HR outcomes or patient outcomes were examined.

Hewison argues that the ideas of evidence-based medicine are widely adduced but remain contested even amongst professional groups (Hewison, 2004). This has implications for its widespread adoption as a form or tool, of leadership given that it requires a leader’s other skills such as effective communication and persuasion rather than displacing them entirely. The processes by which consensus arises within medical practice are not always entirely transparent and much custom and practice still prevails where evidence is lacking. Despite this the actual practices of EBM such as rigorous use of data, willingness to question established practice and recognition of the need to persuade are all skills that can be located in effective leadership practice itself.

20 21

Leadership Styles and Competencies

Authenticity in leadership styles has received considerable academic attention (Avolio and Gardner 2005). Leadership theory stems from the idea that subordinates or followers show more inclination to accept the influence of people they see as enacting ideas and policies in which the leader themselves actually believes. Leadership development that incorporates techniques to bring leaders closer to authenticity have been popular. There is only limited evidence of the effectiveness of authentic leadership styles beyond self-evaluations. Shirey interviewed senior nurse managers in a number of US hospitals to better understand their role as leaders and how they fitted within their various organisations (Shirey, 2009). Despite sharing many leadership attributes and attitudes there was a pattern of increasing barriers to exercising “authentic” leadership in organisations with poor existing cultures mirroring at a higher level, the findings of other observation studies (e.g. Martin and Waring, 2013) describing significant barriers to the exercise of leadership despite considerable individual knowledge of what it “should” be.

Wong and Laschinger surveyed 280 Canadian nurses to better understand the relationship between authentic leadership and nurse empowerment (Wong and Laschinger 2013). The nurses who reported that their leaders displayed elements of authenticity such as openness, self-awareness and adherence to high ethical standards were more likely to report feeling empowered and more satisfied with their work and their own performance compared to others. Wong and Giallonardo (2013) reported lower rates of adverse events for their patients from the same data.

Wallis and Kennedy (2013) examined the effect of a leadership intervention to improve the quality of team work of senior staff aimed at addressing problems of retention of junior nursing staff within their respective organisations. Researchers used measures of leadership practices and emotional intelligence to rate teams’ progress over a year-long development programme. The authors used three existing scales to measure aspects of emotional intelligence. They define the concept as “improving understanding of one’s own emotions and those of others in order to achieve greater self-awareness and more effective social management, especially with a diverse workforce” (p65). The intervention consisted of team away days and individual leadership coaching and feedback. Two of the teams developed good working relationships during the course of the programme. Dysfunction at the group level was seen to feed directly into the relative success for each group. In each of the poorly performing groups the predominant existing leadership style was observed to have a negative impact on collective effectiveness. Change in leadership behaviours was only seen in the groups with good dynamics initially and deterioration was noted in groups where poor leadership behaviour was manifest at the outset. This suggests that even in the context of a leadership development programme, poor leadership behaviour can persist and reduce effectiveness.

Shah et al. (2013) looked at the safety effects of leadership twinning for newly qualified specialist surgeons. No clinical or safety outcomes are reported but anecdotal reports from the participants suggest that the consultants found the course useful at a time of significant personal and professional challenge. Although nurses and other staff are mentioned as being “involved” in the course the article says nothing about multi-professional leadership approaches nor followership. The primary focus is on “coping skills” and professional accountability of the individual rather than the fact of assuming leadership of a multidisciplinary team explicitly. The inclusion of other professional groups

reflects the reality of role for which the consultants are being prepared but leading such a team and how this is most effectively done is not reported as a central aim of the programme. This is a considerable weakness compared to more interdisciplinary models.

Corrigan studied mental health team leaders, members and service users to understand the contribution of leadership style of patient experience of services (Corrigan et al. 2000). Service users preferred teams led by senior staff employing “transformational” or “transactional” styles. Leaders rated as using a “laissez-faire” approach were less highly rated. Subordinate rating of their leaders was predictive of service user rated quality of care in terms of “satisfaction” with care and their self-reported “quality of life”.

In their review of the evidence for leadership development West and colleagues examine the concept of leadership as consisting of a set of competencies (West et al. 2015). They note that there is a profusion of such frameworks but that this has not led to their validation against patient and organisational outcomes although some frameworks have been widely adopted and well received by recipients. The same report includes a summary of elements of successful leadership interventions adapted from Yukl (2013). These include many generic components of good educational materials such as clarity of objectives and “meaningful content”. In addition they suggest that appropriate sequencing of material, a mixture of training methods, efforts to build self-confidence and opportunities to apply learning all contribute to effectiveness.

Personal qualitiesLeadership training can focus on achieving self-awareness and understanding of “natural” leadership styles and personal traits. Recently there has been more emphasis on choice of which personal attributes are useful and which could or should be suppressed. This has become especially important with the rise of collective and system leadership thinking.

Traditional “figure-head” leadership paradigms lean heavily on the importance of individual personal qualities. Such theories have roots in “virtue ethics”, the degree to which good conduct follows from personal attributes rather than merely a sense of duty (to an employer) or rational appraisal of consequences (role as manager). Development programmes can help leaders to derive meaning from their own “back-story” nurturing latent sources of leadership behaviours by promoting reflection (Day et al. 2014). Incorporating such “hermeneutic” or meaning-finding into life-story approaches has been promoted by Sparrowe (2005). The acquisition of personal meaning and the ability continuously to reinterpret experience are seen as key abilities for the exercise of leadership. Such attributes are most valuable in leaders facing more complex or “wicked” problems where adherence to principles and resilience over the longer term to deliver strategic or transformative change is important. These are typical of healthcare and so authentic leadership still retains some appeal. Ligon et al. (2008) suggests that hermeneutic assimilation of early experiences which foster empathy and positive and optimistic attitudes toward people, are important in the development of high-performing leaders. This suggests that development initiatives should draw out latent qualities in candidates rather than try to “teach” important personal qualities.

22 23

Longitudinal approaches to development are also important as skills and relationships can and do develop over time (Day et al. 2014). One study of a 2-year leadership training programme used feedback from subordinate employees to understand the impact of the programme (Palm et al. 2015). The programme included group reflection and a series of seminars on aspects of leadership. Feedback from employee interviews suggested that the leaders had developed in confidence in their role and were more supportive and approachable but as with most such studies it was unclear what effects were directly attributable to the intervention and what could be explained by maturation of relationships that would have occurred anyway. Baker uses a case-study approach to examine high-performing health organisations across a number of countries who were nominated by health improvement experts (Baker, 2011). Consistent leadership was identified as a key common theme across these successful organisations. Other features of effective organisations were their explicit commitment to distributed and collective leadership models and devolution of decision-making to the lowest possible levels. High-performing organisation expended resources on development opportunities for staff. Advancement was made dependent on training and support with a primary focus on direct skills and knowledge for quality improvement and measurement. Such skills were viewed as core components and the foundations of leadership across these organisations. For example, in one organisation (Jönköping County Council in southern Sweden with responsibility for public health and community care) over half of all staff had had training at the local quality improvement training “hub”. The ability to find and use information was also seen as central to leadership development in renowned organisations such as Intermountain and the Veterans Health Administration in the US. Topic experts were frequently put in place to lead the clinical decision support and pathway development projects used continuously to improve patient outcomes. Each organisation employed the same principle, essentially that distributed leadership meant those with expertise were expected and supported to lead. An important role for more senior leaders was seen as developing leaders at the “Microsystems” (ward or department) level as a way to tie in all levels of the organisation together to deliver top to toe quality improvement.

Both Myers-Briggs questionnaires and models drawing on the five-factor model of personality (neuroticism, extraversion, openness, conscientiousness, and agreeableness) are commonly used (McKimm and O’Sullivan, 2013). These tend to be valued by participants but there is little evidence linking them to objective outcomes. Such “person-role merger” is important for creating “authentic” leaders and the use of self-generated narratives regarding formative experiences can help leaders understand their own development and the importance of their own personal journey. The emotional and communication skills required to deliver system or distributed leadership may benefit from such introspection amongst other strategies. Leadership development can draw attention to and give space to reflect, on these sources of professional effectiveness.

Other research suggests that the effectiveness of leadership cannot rely solely on “soft” skills and that flexibility and adaptability of leadership styles is key to dealing with underperformance and not just pleasant situations such as offering development opportunities. The Hay Group published a report of observational research conducted in a number of settings entitled “Nurse Leadership: Being nice is not enough. A Research Study Examining the Value, Impact and Leadership of the Ward Manager” (The Hay Group, 2006). The research showed that some nurse leaders were able to work with staff in a supportive manner but were less able to deal with substandard care and capability management. They also found that relying of cooperative management styles alone could lead to resentment as high performing staff felt unsupported by leaders’ inability to challenge poorly performing colleagues. This emphasises the point made earlier about the requirement for style switching to suit the situation or even individual colleague.

Primary Care and Community Leadership Primary care is at the heart of the NHS and it retains a pivotal role today as evidenced by the leading role taken by Clinical Commissioning Groups in shaping NHS spending patterns. Unfortunately the leadership strategies or abilities that best support clinical quality improvement leadership in primary care has received little academic attention. Chreim studied distributed leadership in the context of change agency across primary care systems in Canada (Chreim et al. 2010). The same markers of high-quality leadership emerged in this setting as in larger organisations. Important factors were leaders’ abilities in finding common ground across organisations and establishing trusting, credible and lasting relationships. The needs of primary care argued for the use of non-traditional forms of leadership such as “distributed” and “system” models. The authors describe distributed leadership as operating when “legitimacy, authority, resources [and influence] are dispersed across loci” (p.187) yet still suggests that an identified leader or groups of leaders are required whose “main responsibility” is to drive change given the time restraints on practitioners. Other authors suggest by contrast that such system leadership across smaller entities is often undertaken by those for whom a change isn’t their primary responsibility. Timmins goes further and claims that effective systemic change requires no fixed leadership to achieve alignment of objectives in complex interlocking networks of agencies. Although increased experience and consistency of senior leadership are often cited as important factors contributing to success they are insufficient without mobilisation of grass roots staff to actually cross boundaries and do joined up work (Timmins, 2015).

Smith and colleagues examined leadership in Primary Care looking in detail at specific practices over time (Smith et al. 2013). The report provides two case studies of expanding Primary Care organisations and how they develop from collections of individual practices. They suggest 12 “design principles” of use to leaders which underpin effective change planning. Leaders in both organisations provided “management” training at a variety of levels for a number of professional groups. The authors note that governance and decision-making models changed over time to cope with the expansion in services and geographical reach of the practices studied. Smith argues that distributed leadership is essential to allow expansion and diversification in primary care given the clinical workload and other pressures already impacting on key individuals. A further requirement is to give such leaders time to reflect on their achievements and to imagine new models and systems of care. At the same time they suggest that individual GPs could take on specific leadership roles rather than attempt to lead as “an add-on the to the clinical day job” (p.54). The report notes the difficulty in accessing high-quality leadership development and support in Primary Care given the diffuse nature of the workforce.

24 25

Little research examining leadership community and nursing home settings was found. One observational study examined leadership in community nursing teams in the UK (Cameron et al. 2012). They found that junior staff rated leaders more highly who they perceived as valuing their contribution and abilities as practitioners. They also found that the senior nurses observed and interviewed displayed “transformational leadership” qualities in tackling the difficulties they faced. A recent report by the University of York for the Royal College of Nursing Foundation examined the delivery of nursing care by registered nurses in nursing homes in England (Spilsbury et al. 2015). Registered nurses (RNs) in this setting coordinate and supervise a much larger workforce largely consisting of untrained carers. Such staff supervise and directly deliver care, to half a million people across 12,000 homes with more beds in total than the NHS. RNs are either directly employed or as NHS community staff but all provide leadership and expertise to a non-professional care workforce with minimal training. This places them instantly in a leadership role whatever their previous preparation and support for enacting clinical leadership. They found high turnover and low evaluation of nursing home careers as significant barriers to the exercise of leadership. In addition participants expressed concern that the whole nursing and care home sector was undervalued by the NHS and that the provision of career development including leadership development could have increasingly important positive impacts for the NHS. Clearly high cost bespoke leadership development is neither possible nor warranted using current modes of delivery. However, there is a clear need for the development of suitable leadership training materials to ensure that such staff understand the influence they have and can monitor, maintain and improve the quality of care delivered by their teams.

Community leadership roles may have significant implications for the wider NHS in terms of the potential to reduce unnecessary admissions and poor care. We examine the extent to which existing nurse training provides leadership development in more detail below. The academic evidence suggests that existing nurse training does not adequately prepare nurses for leadership roles and that nurses themselves are aware of this shortfall. Even less is known about the training and development of care home managers. Spilsbury et al’s (2015) study raises a number of important issues for leadership development. The authors recommend research into the impact of quality of leadership on quality of care in nursing homes given its likely impact on the NHS. They ask what qualities are most needed to lead within the care home system and what impact does quality of leadership have on staff well-being, burnout and turnover. This area of healthcare delivery has been almost entirely neglected in leadership research. This review could find no research focussed on the importance of the size, training or composition of the nursing home staff being led by RNs nor the level of seniority or experience of the RNs themselves. There was no research identifying leadership capabilities or development interventions designed to maintain and promote quality of care in isolated, small team or independent practitioner settings such as nursing homes despite their increasing importance for the UK healthcare system as a whole.

The fundamental driver to improve and spread leadership in the NHS is the need to improve services whether by using research evidence, improving efficiency or responding to failures in the safety or acceptability of care. Policy documents and other sources have often implied that individual leaders are expected to influence organisational performance and health outcomes directly. In fact it is increasingly accepted by leadership academics that leadership development can achieve such effects only indirectly (Day et al. 2014). Instead it is more useful to measure the effect of leadership development on organisational culture or other intermediate indicators such as morale or improved communication thought likely to influence performance. This is because of the sheer complexity of healthcare organisations and the fact that many other factors also contribute to success or failure sometimes in ways that over power the impact on leaders. The following section explores the concept of measuring leadership styles, the academic and non-academic measurement of the impact of different leadership interventions, leadership for engagement and cost-effectiveness of leadership interventions.

Measuring leadership stylesEmphasis on the abilities of individual leaders and the need for suitably directive styles of leadership have seemed a natural fit and been amplified in the rhetoric. But there is increasing evidence that top-down and directive styles of leadership may be ineffective or even counterproductive. A large survey of 800 NHS senior leaders found most used “pace-setting” styles where “stretch targets” and direct control were seen as the currency of success. These styles were more often seen as demotivating to professional followers who make up the majority of key staff within the healthcare industry. Treating such staff as tools with which to enact high-level goals and meet targets may have acted to depress organisational performance in many cases (Santry, 2011). West reviewed evidence relating organisational culture and “good performance” including safety and concluded that “dominant hierarchical cultures” could have a negative effect on patient safety (West et al. 2015). Coaching was reported as effective for more senior learners but there was no evidence to suggest longer programmes were more effective.

Careau et al. (2014) conducted a systematic review of leadership programmes and found that four in five reported some form of evaluation. Most studies demonstrated impacts on learners only with less than 7% and 4% presenting evidence of patient impact or system changes respectively. The large variety of measures and competencies reported meant that the results could not be pooled to show important impacts across the body of research. Aarons et al. (2014), for example, has developed a scale to measure “implementation leadership” which can be used to understand the effectiveness of leadership specifically aimed at implementing evidence based practices (Aarons et al. 2014). The dimensions describing implementation leadership that emerged from the data as important in demonstrating impact of leadership development were: knowledge, proactivity, supportiveness and perseverance. Leadership development that incorporated such elements could then be tracked through trainees self-reports of achievements, the assessment of mentors and on to actual implementation projects. These findings echo those of Avolio in another review who found limited outcome data on organisational performance but much more on “affective, behavioural and cognitive” outcomes i.e. measures of individual attributes before and after training (Avolio et al. 2009).

Question Two: Which indicators can be used to assess the impact of Leadership development programmes?

26 27

Another component of leadership is the need to develop and communicate guiding visions of where a service should be aiming and what it should be achieving. Such skills are key to designing and creating transformed services to deal with established problems in novel ways but such skills are difficult to evaluate. Martin et al. (2012) evaluated a senior nurse leadership development programme focussed on helping them to generate effective and compelling visions for their teams and organisations. The intervention was modelled on the RCN Clinical Leadership Programme and evaluated in Switzerland. The study found that the development of strategic vision required dedicated time, reflection and contemplation. Some participants expressed the work as analytic in nature. For others it was more akin to a creative and intuitive process. Most respondents agreed in contrasting “visioning” (sic) to other leadership and managerial tasks they were accustomed to completing and valued the programme for increasing their effectiveness in this area of leadership.

Enterkin and colleagues evaluated a leadership programme for future ward sisters. Approximately 60% (n=60) of three cohorts responded to an invite to complete a semi-structured questionnaire from which qualitative data were extracted and analysed thematically (Enterkin et al. 2013). The programme consisted of “taught workshops” and action learning opportunities. Respondents noted an increased understanding of their potential role in empowering colleagues. Many also noted that the programme increased their understanding of their organisation, their role within it and a greater appreciation that the challenges they faced were similar to those of colleagues in other areas. One common personal difficulty was in organising opportunities to shadow seniors (mentorship) with only a “small number” able to experience this. The “academic” components of the course were also experienced negatively by some but action learning sets were viewed positively as a rare chance to reflect away from day-to-day pressures. What was positive was that the findings from respondents were used to adapt the programme and resource constraints were noted such as the inability to offer 360o feedback opportunities to all participants.

Cummings conducted a systematic review of nursing leadership research including papers which evaluated the effects of leadership development interventions (Cummings et al. 2008). Nine uncontrolled before-and-after studies were described. All found that leadership development led to increases in leadership behaviours as assessed by self-reports and independent observations up to 18 months afterwards. The interventions varied greatly in length (from 3 days to 18-months) and format (self-directed learning, workshops and residential courses) but the paper did not report if these differences were reflected in the degree or types of improvements seen.

Duffy and Carlin (2014) depict a development programme for Band 6 (“Junior Sister”) nurses. The authors used a four level evaluative framework developed by Kirkpatrick (1994) to categorise different items and outcomes included in the course. “Reaction” items included relevance, novelty and ranking of individual content. “Learning” items were assessed via email contact at six months with some respondents reporting that they had not achieved some of the goals they set themselves during the programme. The primary reason suggested was the variation in “learning cultures” candidates experienced on their return to practice. “Behaviour” items were assessed using feedback from the learners’ line managers although only anecdotal evidence was reported and the rigour of this process was unclear. Finally the authors concluded that the “Results” of the programme were difficult to measure and although they suggested that student, line manager and patient feedback data

alongside nursing dashboards could be used they did not report any findings. Such data could be difficult to use given the many alternative explanations for changes in such indicators although before and after measures of more structured metrics or rigorously analysed interview data would provide more robust information.

The use of control groups is rare in the leadership research literature but would significantly increase the validity of leadership development evaluations. Biggs set out to evaluate a leadership programme for senior police officers using subordinates’ feedback of those receiving and not receiving (controls) the intervention at baseline and seven months after the intervention (Biggs et al. 2014). They noted that much previous work had centred on rectifying negative outcomes. In contrast they focused on how the leadership intervention increased organisational strengths and enhanced subordinates’ positive experiences and engagement. Such outcomes are often amongst the stated goals of leadership interventions but rigorous measurement is often lacking in published evaluations. The public sector police service settings shares much in common with healthcare in terms of the stressors such as unpredictability, shift working and the need to cope with emotionally challenging work. The intervention consisted of 360o feedback, action learning sets and individual coaching of an intensity similar to many reported programmes in healthcare. Various psychological scales measuring aspects of stress and wellbeing were measured along with turnover intention. Intervention participants reported similar or worse measures at baseline than controls.

Compared to controls, improvements from baseline for intervention subordinates were found in reported “work-culture support, strategic alignment, work engagement, and job satisfaction” (p. 47). These are all important aims for leadership interventions aimed at improving the intermediate outcomes of workplace wellbeing. No significant effects were found on responses measuring job demands, supportive leadership, turnover intentions, or psychological strain. No similarly well-designed evaluations of leadership programme impacts on the welfare of followers were found for healthcare workers within the NHS during this review.

Rosenman et al. (2014) conducted a review of research examining the use of “immersive simulation” of team and work environments to increase “team leadership” in healthcare settings and found 42 relevant studies. They found that simulation-based interventions only appeared in the literature from 2006 onwards. Most studies focussed on complete teams, only four focussed mainly on leadership and only one study assessed leadership as a primary outcome. Patient outcomes or direct observations were used to evaluate the interventions in only four of the studies included. Rosenman et al. (2015) also reviewed the literature examining the types of tools that have been used to assess leadership in such circumstances. Two-thirds of the papers studied leadership at the team level with the remaining third focussed on individual leadership behaviours and skills in directing team performance. There was considerable variation in the evaluations including knowledge and skills tests, and observations of behaviour changes. Much of the research in this area falls under the label of “human factors” despite considerable overlap with leadership.

28 29

Leaders’ reactions to feedback is thought to be important there is no agreement as to how this should be evaluated or consistent interpretation of the effects of feedback on future performance in leadership positions (Day et al. 2014). One problem with measuring leadership in terms of feedback is the relative level of the source. A subject’s subordinates value different qualities from their superiors and possibly few leaders are evaluated as effective by all levels simultaneously. A further problem is that the quality of feedback is important. Training for raters or even payment, can both increase the quality of feedback and consequentially its usefulness. Day and colleagues conclude that leadership development programmes can be evaluated by measuring changes in social networks after interventions, using Q-Methodology to reduce variable responses to manageable themes and models. Calculating return on investment has also been tried but there is little consensus over which method achieve valid comparisons across vastly differing organisational contexts (Day et al. 2014).

Day argues (p.77) against measuring job performance as a proxy for leadership given the risk of confounding by other factors as what is measured might be organisational culture rather than leadership itself. Other papers suggest that intermediate outcomes such as turnover intention, whilst still biased by institution level factors, do appear to be more directly related to quality of leadership than high-level outcomes. Li et al. (2011) compared leadership measured as a label (yes/no) and as the number of subordinates with various personal attributes and abilities. They found no correlation between being in a leadership role and intelligence or academic achievement. They did note a positive relationship with self-esteem and self-efficacy. Women from higher socio-economic backgrounds enjoyed less success in attaining leadership positions overall compared to men after adjustment for education and other factors. This has important implication for evaluating programmes themselves given the gendered differences in barriers.

Measuring impacts of different leadership interventionsIn future, innovation in the NHS will dominated by efforts to achieve integration of care services along patient journeys and pathways and therefore across organisational boundaries (Storey and Holti, 2013b). Systems leadership has been defined as “seeking to make change across organisations where people did not have a direct, line management responsibility” (Timmins, 2015:7). Initiatives to develop and support the requirements of “system leadership” may present greater challenges to evaluation than more traditional models. Recent work based on interviews with prominent people who have acted in leadership roles in such circumstances suggest that the move away from heroic leadership in this context is even more profound (Timmins, 2015). Many of those interviewed claimed not to be in any kind of systemic leadership role but rather to exert “soft power” and work “behind the scenes”.

The origins of system leadership thinking lie in the increasing interdependence of modern life and the complexity of the problems it throws up. Despite there being an apparently ever increasing multitude of organisations set-up to tackle various problems and deliver a variety of services the problems they address and the solutions they require are increasingly interconnected. It is to this dilemma that system leadership is a proposed remedy. System leader is a term applied to individuals and is not used to refer to the processes of collective or distributed leadership which are situated in the relationships between various agents. What is thought to characterise such leadership is the willingness to cross boundaries in search of solutions.

Senge suggests that system leaders require three abilities above all others:

n To see the system in its entirety

n To create thinking spaces to enable reflection

n To turn the “collective” focus towards “co-creating” a future

(Senge et al. 2015)

Whatever model or structure of system leadership is chosen, the ability of an organisation to realise these three central aims is key wherever “wicked” problems of making separate services feel like a pathway for patients and families (The Kings Fund, 2012a) is the issue. Evaluation of the impact of system leadership development is in its infancy and faces greater methodological hurdles given that system leadership is hard to identify let alone measure.

Non-Academic Measurement of Leadership and its impacts One of the five questions the Care Quality Commission (CQC) now asks of all providers is “how well-led is this organisation?”. The commission suggests that “the quality of leadership is one of the most important determinants of the quality and safety of services” (Care Quality Commission, 2014a). The reinvigorated inspection regime has placed evaluation of organisational leadership centre stage in its assessments (Care Quality Commission, 2016). The Commission now uses a variety of direct inspections, staff and patient interviews, walk-arounds and documentary and data analyses to evaluate the impact of leadership amongst other things.

“Professor Sir Mike Richards, CQC’s Chief Inspector of Hospitals, said: ‘The leadership of a trust sets the tone for the organisation. It’s a vital area for all three regulators to look at, so it’s only right that we should have a shared view of what characteristics a well-led organisation displays” (Care Quality Commission, 2014b) This view is challenged by much of the latest thinking regarding leadership. The idea that the top of an organisation is the only or even the main place to look for leadership is on the wane (The Kings Fund 2011 and The Kings Fund 2012a). The rise of followership and an increasing recognition that “you can achieve almost anything so long as you don’t want to take credit for it” (Timmins, 2015:8) make the case for leadership across boundaries compelling, but harder to define and measure. If by definition system leaders have no or little traditional authority the challenge is what should they use instead and what should leadership development get them to concentrate on.

30 31

Systems leadership relies on some new ways of thinking about leadership and its components:

n Evidence – not “do as I say” but “do what we know works”

n Connectivity – know who is who and where they can be found

n Understanding – A birds eye view of the whole system is needed to convince others that proposals will work across the system and not just in one patch or part of a network of care

n Influence – just as much leadership and followership is defined in terms of ability to influence others this is one area of systems leadership that remains an essential ability to effect change

n Responsibility – it seems necessary for system leaders to “feel responsibility” rather than formally having it.

n Ownership – cultivate the ability to give ownership away

(Adapted from Timmins, 2015).

The Care Quality Commission has also paid significant attentions to the quality of leadership in organisations as a marker of high quality. “By ‘well-led’ we mean that the leadership, management and governance of the organisation assure the delivery of high-quality care for patients, supports learning and innovation and promote an open and fair culture.” (CQC, 2014b:1). The CQC provider handbook for acute hospitals sets out ways in assessing whether and organisation is “well led” is undertaken and identifies it as one of the 5 “key lines of enquiry”. The CQC report, “The state of health care and adult social care in England 2014-15”, found that quality of care was almost invariably associated with high quality leadership: in 94% of organisations rated “good or outstanding” for care overall received the same rating for their leadership. The converse was also found: in 84% of services rated “inadequate overall” inadequate leadership was also identified (CQC, 2015:8).

The report also synthesises information from inspection reports, focus groups of inspectors and senior staff to assess the role of leadership in delivery safe, high quality care. They identified five aspects of leadership of “critical” importance:

n Effective engagement and communication with staff and service users

n Skills, experience and “visibility”

n Strong and positive culture within the organisation

n Learning from mistakes

n Open and transparent governance

In addition the report emphasises three areas of leadership that directly increased quality:

n Leaders using engagement to build a shared ownership of quality and safety

n Staff planning that goes beyond simple numbers and includes skill mix, deployment, support and staff development

n Working together to address cross-sector priorities

For the purposes of this review, these latter three priorities can be thought of as “distributed leadership”, “followership qualities” and “system leadership” respectively. This demonstrates that the CQC inspection regime for health and social care has accommodated the need for new concepts of leadership and evaluation, to support quality improvement as new models of care are constructed.

Methods of evaluating leadership have also focussed on specific groups of staff which reflects that the experience of leadership can radically differ across professional and managerial groups. Elliott and colleagues developed a set of leadership “outcome-indicators” to allow demonstration of leadership by advanced nursing practitioners (ANPs) (Elliott et al. 2014). The aim was to capture the impact of the considerable amount of leadership activity that many ANPs undertake. The report finds that a majority of ANPs undertake aspects of leadership within their roles and spend around 15% of their time on “coaching, teaching, protocol development and research” all of which can be thought of as leadership roles with potential effects on the quality of patient care. The outcomes indicators were categorised into four groups: building capacity and capability within multidisciplinary teams; measures of esteem; innovation in clinical practice and service delivery; evidence-based practice. They argue that such outcomes should be included in evaluation of leadership development programmes with aspects such as self-esteem being of particular importance for professions with lower levels of instrumental authority. Evaluation of ANPs and other emerging professional strata such as physio-practitioners will be of increasing importance in mapping the impact of leadership development as these groups will increasingly take on responsibility for guiding their much more numerous subordinates, all undertaking direct care roles with significant potential impacts on patient outcomes.

Leadership for EngagementThe degree to which staff are positively engaged in their work is associated with many markers of health care quality and so measuring this concept could provide a useful way of assessing the effect leadership development interventions (The Kings Fund, 2012a).

The “Index of Medical Engagement” is a set of leadership indicators specific to healthcare contexts. These scales were developed by the NHS III, The Academy of Medical Royal Colleges and Applied Research Ltd. The indicators are divided into three domains: working in an “open culture”; having purpose and direction: feeling valued and empowered. The validation study found that the scales correlate well with performance data including safety in a study of 30 NHS trusts and as such is one of the few tools available to link leadership development to performance in the UK health care setting.

32 33

The use of NHS staff survey results combined with Trust level outcomes (Powell et al. 2014 and West et al. 2011), could be adapted to allow longitudinal analysis. More robust research would then be possible if discrete leadership interventions could be identified and suitable comparator sites without significant leadership development initiatives were available to look for a directional effect. Data sets from the NHS itself rather than inference for other contexts such as commercial, non-health or non-UK healthcare settings could be useful in demonstrating the effectiveness leadership development before and after. There is some evidence measuring links between “staff engagement” and performance within the NHS (King’s Fund, 2012a). Engagement is a key aim of effective leadership especially for shared or collective leadership. Without such evidence, it is difficult to demonstrate the degree to which leadership development achieves its aims. Evidence such as Powell and Dawson above suggests that outcomes such as absenteeism and turnover do affect patient care and can themselves be improved by good leadership. It is certainly rational to suggest that lower absenteeism and turnover lead to higher quality care and lower costs (West et al. 2011). McLeod and Clarke (in Powell et al. 2014 p13) examined employee engagement across a range of organisations in the public and private sectors (non-healthcare). They found a “moderate” correlation between engagement and improved performance suggesting that leadership development which enhances engagement could play a role in driving improvements.

Cost–effectiveness of leadership developmentAvolio and colleagues show that return on investment can vary from negative to 200% depending on a number of factors (Avolio et al. 2010). The calculations rely on assumptions about the “dollar value” of improvements in performance and effect sizes and duration of effects are specific to each programme. It is unlikely that such methods can easily be applied within the NHS although generally positive cost-benefit ratios suggest that such training is rarely a waste of money. They suggest that there is broad evidence for some positive impact on organisational performance but argue that lack of cost-benefit data lead to development programmes being thought of as “optional extras”.

There is evidence that leaders can enhance team performance (Nash and Govier, 2009). Team reflection at meetings for example can help to improve team working but to assist the emergence of more effective team working, leaders have should take a role, understanding that team working can be improved and that they can act as facilitator once they understand the members and dynamics.

Many evaluative strategies have been adopted to measure the impact of leadership programmes but in healthcare contexts links with high-level performance are hard to justify. Measuring the impacts on intermediate outcomes is a more robust and fruitful approach and remains an improvement on evaluations replying exclusively on self-reported data from participants.

The emerging need for system, collective and distributed models of leadership to deliver integrated healthcare implies that previous uni-professional development programmes could be less effective for a transformed NHS (The Kings Fund 2012). Counter to this, is the concern that the existing hierarchies within healthcare may disadvantage certain groups and professions in mixed leadership training initiatives. It has been argued that the role of “first level” leaders is “critical in organisational effectiveness” (Aarons et al. 2014) whereas as other authors have focussed on the importance of the high-level leadership function of nurturing suitable organisational “cultures” (Davies et al. 2007 and Dixon-Woods et al. 2014).

The following sections examine the evidence for the relative effectiveness of leadership development focussed on single professions, specific settings, across professional boundaries and at differing career stages.

Nursing leadership Wong et al. (2013) reviewed the evidence for clinical and managerial leadership within nursing. Counterintuitively, the authors situate leadership within Donabedian’s ‘structure- process-outcome’ framework as a structural component in terms of delivering quality of care. The authors found a small number of papers connecting high quality nursing with “relational”, as opposed to “task-oriented” leadership. They found evidence that improved outcomes such as lower mortality, improved drug safety, patient satisfaction, lower nosocomial infections and reduced use of restraint techniques were seen where leaders displayed emotional and communicative sophistication in guiding junior staff at the bedside.

Maintaining such improvements can be challenging where workforce pressures reduce the availability of suitably experienced and skilled staff. Griffith examined the issue of succession planning to maintain high quality nursing leadership in the context of global shortages of nurses (Griffith, 2012). The author suggests that the connection between adequate nurse staffing levels and safety can often be attributed to a lack of leadership “at the bed side” and at more senior levels. The available literature suggests that succession planning in nursing is neglected by many healthcare organisations (Titzer et al. 2013). Where programmes were identified, little evidence in terms of patient outcomes or nurse turnover was found to demonstrate the effectiveness of existing initiatives. The Care Quality Commission also emphasise that consistency of leadership is important at all levels and that planning for succession is an integral step in maintaining a focus on quality over time (Care Quality Commission, 2015). This suggests that leadership development programmes for nurses should include emphasis on “passing the baton” as a key responsibility of good leaders.

Tregunno examined Intensive care to understand the role of nursing leadership in increasing patient safety (Tregunno et al. 2009). They found that leadership “at the bedside” appears to have the greatest impact on safety given the need to react effectively to changing circumstances. But this ignores the widespread impact of higher level initiatives such as the introduction of care bundles which are often negotiated at a higher and more traditional level of leadership by senior nurses and medical staff. In addition it has been the case that leadership development has not been focussed on bedside nurses or other junior staff. Innovative online and self-directed leadership training may be needed to widen access amongst clinical staff at a realistic cost. Enabling hand-on nurses to enact leadership behaviours is a key feature of shared-governance models to encourage ownership of patient safety and quality of care by those with most direct impact on it.

Question Three: Can leadership development be better targeted? Is there evidence of impact for single or multi-professional groups, in certain settings or at specific levels of seniority?

34 35

Fealy et al. (2011) conducted a large survey of nurses in Ireland and found similar problems relating to power and influence. Lower grade nurses were generally restricted to demonstrating high levels of clinical competence to influence others by example rather than exerting traditional leadership in many cases. Higher grade nurses felt more able to influence the clinical work of others but even here not all hierarchies were found to be equal within the organisation with medical leaders thought to have relatively greater influence. Interestingly access to leadership development was associated with a reduction in the feeling of “lack of recognition”. Casey et al. (2011) examined the leadership development needs of nurses and midwives and stressed that differing grades expressed different needs and that workplace interventions were seen as particularly valuable. In a similar study employing participant self-reports and acceptability only long term and workplace-based mentoring and coaching was highly valued (McNamara et al. 2014).

Evidence-based practice councils in nursing have been evaluated for their potential to develop leadership qualities in nursing staff in addition to directly improving job satisfaction and quality of care (Brody et al. 2012). Participants reported feeling greater empowerment, they valued opportunities to exercise leadership and gained from exposure to quality improvement methods. This underlines the need to see leadership development and quality improvement as mutually reinforcing.

Medical Leadership The current state of medical leadership within the NHS has been explored relatively recently using survey and case study methods (Dickinson et al. 2013). Medical engagement measured using the Medical Engagement Scale, was positively related to some measures of improved performance. These indicators included both self-reported survey results and publicly available indicators such as CQC inspection findings and Standardised Mortality Ratios. As with the vast majority of such evidence it is unclear whether medical engagement is a cause or a result of better than average performance. The authors also identified “gaps” between medical leaders and their colleagues in many Trusts which confirms evidence from other studies describing similar professional barriers to assuming leadership roles for medical staff specifically.

As noted above, leadership for clinical improvement and innovation in healthcare is frequently difficult or even impossible without medical engagement (Bohmer, 2012 and Bohmer, 2013). A common strategy to remedy poor engagement has been to invoke the concept of medical leadership. Doctors are conceived as central to clinical “Microsystems”, patient care delivery and high level arguments over resources and are well placed to understand patient impacts and outcomes (Darzi et al. 2008). There is considerable evidence that the more involved “engaged” clinicians become, the better the outcomes that can be achieved (Ham and Dickinson, 2008). Despite this some authors have noted that there are high levels of suspicion about managerial motives amongst many doctors and this is greater at the “front-line” (Martin and Learmonth, 2012 and Bohmer, 2013). Exhorting leadership from doctors has often been seen as a tactic to smuggle in managerial objectives by granting a veneer of clinical control or by aligning the aims of influential medical staff with political or executive imperatives. International examples of clinical leadership in high-performing organisation discussed above do not bear out these suspicions in the main (Ham and Dickinson 2008 Baker 2011).

Recent policy has re-emphasised the importance (and current variability) of clinical involvement with the advent of Clinical Commissioning Groups with large-scale budgetary control (Naylor et al. 2013). The degree to which these bodies are yet to transform levels of clinical leadership is contested but such organisations are structured in such a way to ensure clinicians’ viewpoints are represented. Patient leadership will be the next step in developing services that fit user priorities (The Kings Fund 2012b). In each case the same challenges of preparing participants to achieve their potential argue for tailored leadership development initiatives (Giddings and Williamson 2007).

Storey and Holti examined clinical leadership in achieving cross-boundary service redesign (Storey and Holti, 2012). Clinical leadership was exercised at “a number of interlocking levels” and required close collaboration across professional boundaries where by definition a clinical leader from one discipline has less exclusive and specialist knowledge. In such cases more generic leadership training may provide the required interpersonal skills such as “micro-political capabilities”, correctly identifying and negotiating with multiple stakeholders and remaking the case for their own clinical role to better fit the priorities of others.

Rising demand, complexity, “success” targets and the commissioning agenda all require at least some specifically medical leadership to make progress possible. Bohmer argues that some of these developments have led to medical disengagement especially from target and performance management (Bohmer, 2012). A role for leadership development therefore is to reaffirm the unique contribution that medical leadership can make and provide potential leaders with the skills to understand the new world they find themselves in once they move even a little way from their patient care comfort zone. Bohmer suggests that doctors should come to think of clinical leadership in terms of high-quality consultation. When dealing with individual patients they ask, “What are your ideas, concerns and expectations?” (p.20). Adopting this approach in a leadership role is not only consistent with the emerging ethos of collective leadership but is instantly recognisable as a core mode of working for effective clinicians and thus a “transferable” skill and source of authenticity.

In medicine there has long been a tradition of clinical leadership although in recent decades the profession itself has tended to feel it has ceded managerial authority to non-clinical leaders (Ham, 2012). Alongside this is a perception that as a professional group, doctors have lost influence over and responsibility for, quality at an organisational level. More recently a return to the importance of clinical leadership has been seen in policy documents if less so in research literature. The evidence from studies of senior medical and managerial staff attitudes and experiences suggests that both sides locate such influence and responsibility in the opposing group. Joint leadership training could play a role in reducing these misunderstandings.

A key objective of future leadership development needs to be the degree to which doctors can maintain the authority and authenticity they derive from providing individual patient care.

Indeed doctors and to a lesser extent other professional groups, look outside of NHS management structures for a large part of their legitimacy and accountability. As Bohmer notes, unlike many commercial organisations, the NHS has limited scope to reward or sanction staff in relation to performance targets.

36 37

It is often argued that doctors have a unique place within healthcare to both effect and obstruct improvement initiatives (Willcocks and Wibberley, 2015; Bohmer, 2012; Chapman et al. 2014; McKimm and Swanwick, 2011 and Stoller, 2014). Evaluation of the leadership abilities of medical staff has been a focus of research attention for many years. Given there is considerable medical engagement in the UK, as members of Trust boards, in Primary Care Trusts (and now CCGs) and as research, directorate and departmental leaders increasing the effectiveness of this cohort, such as providing structured support to develop skills in HR, finance and team leadership has and should continue to be a major priority for the NHS (Stephenson, 2009).

Given their level of autonomy, expertise and their central role in distributing healthcare to patients they are the single most significant “professional bureaucracy” with influence over quality of care (Ham and Dickinson, 2008). Recognition of this fact has led to repeated calls for clinical leadership. But support for doctors moving into such roles can be limited and the rewards nebulous (Bohmer, 2013), although successful examples do exist (Agius et al. 2015; Ham, 2012; Ham and Dickinson, 2008; Smith et al. 2013; Starck and Rooney, 2015; Stephenson, 2009; Storey and Holti, 2013a and Straus et al. 2013). In terms of a policy objective, medical leadership in the NHS has been promoted for longer and more widely than in many of the other countries examined (Ham and Dickinson, 2008).

The existing training of medical staff may fit rather well within the complex world of the health service leader. Bohmer differentiates three modes of care that doctors implicitly understand: routine, repetitive; constrained problem-solving, individualised; unconstrained, complex, experimental, problem-solving (Bohmer, 2012:11). These varying categories of activity map well onto models of innovation (the third level of complexity) and how given changes must integrate with other organisational activities at lower levels. Leadership development can use such familiar conceptualisations to help doctors to understand the kinds of activities that comprise leadership for clinical improvement, not of individual patients but of whole cohorts and along entire pathways. Another natural fit between leadership and medical staff experiences expertise in the choice of goals for their respective organisations. Such choices, when viewed as purely political or managerial, can meet with resistance from clinicians. Clinical leaders may have greater credibility to argue for alignment of professional and managerial goals based on valid evidence.

McKimm and O’Sullivan discuss the personal qualities needed to lead from a medical perspective as highlighted by The Francis Report and other key documents and ask where developmental training can have greatest impact in the integration of technical competencies and leadership skills (McKimm and O’Sullivan, 2013). The authors suggest that many of these qualities cannot be taught formally but could be enhanced in work contexts. To increase practice-based opportunities to develop and demonstrate leadership, development programmes should include long-term mentorship or coaching in a “safe” environment but focussing on real problems and solutions. The authors also discuss the emotional “skills” of potential medical leaders given the existing high emotional work component of their conventional roles. They caution against formal assessment of the results of such programmes until a firmer evidence base for the effects of emotional and personal leadership characteristics is available so that a natural and healthy variation in styles is not “artificially reduced”. Blumenthal discusses the need for “systematic” medical leadership training for “frontline” clinical leaders (Blumenthal et al. 2012). They argue that leadership development should begin in medical school whereas it is currently not emphasised leaving medical staff to develop such skills later in their career.

They note the evidence that good leadership has an impact on patient outcomes (e.g. Wong et al. 2013; Corrigan et al. 2000, and Morrow et al. 2014) and suggest it should receive attention in clinical curricula in the same way that individual knowledge and skills do at present.

Patient Leadership Authenticity in healthcare leadership is often defined in terms of fidelity to patient views and needs. The results of the Francis enquiry demonstrated how leadership can fail in the NHS. Key areas of failure were identified as a lack honesty and denial of poor standards of care by staff when faced with complaints (Francis, 2013). This has led some to call for the establishment of formal leadership roles and development support for patients and families themselves (The Kings Fund, 2012b). The Kings Fund itself has developed a step-by-step guide to “experience-based co-design” adapted from a model to involve end-users in architectural design projects (The Kings Fund, 2013). They argue that leadership should not merely “capture and tame” patients in the name of patient involvement but create real leadership and project development opportunities for them. National Voices have also urged greater investment by the NHS in patient and public leadership (National Voices, 2012). Much of what they advocate is either genuinely new (and therefore untested) or needs radical expansion to have measurable impact. Finally, guidance to local providers on creating Sustainability and Transformation Plans refers to “six key principles” of public involvement (NHS England, 2015). These place a responsibility on the leaders of local health systems to support patients and carers to make meaningful contributions at every stage and level to help deliver the Five Year Forward View. All of these policy initiatives presuppose that there are considerable numbers of patients and public available to take up such roles successfully.

In this context it is disconcerting that very little is currently known about the development needs of patient and public leaders in healthcare. This is perhaps unsurprising if the assumption is that as patients they are already experts by definition and therefore all that is required is for healthcare staff to listen. And yet if expert health care staff require leadership development surely a similar logic applies to patients? A central message of virtually all leadership research is that knowledge and even authenticity count for little without a set of other abilities to enable individuals to exert tangible influence for change.

Evaluations of patient leadership programmes are rare. One report of a pilot project found that patient participants valued the opportunities to network, involvement in Action Learning Sets and individual coaching (Swage, 2014). Overall participants felt that the programme needed to be more responsive to the needs and previous experience of the participants and include more opportunities for activity-based working rather than formal teaching. Coulter describes eight key factors needed to establish real patient leadership in healthcare (Coulter, 2012). In summary Coulter argues that patients need a form of distributed or shared leadership. Such a model should enable patients and families to voice concerns and personal preferences for care at every level of the NHS from individuals to lay membership of boards or national bodies.

38 39

Brett conducted a review of the effects of PPI in research in health and social care (Brett et al. 2014). The review identified a number of areas where PPI had measurable impacts. Usefully the review also identified areas where PPI required greater support to have influence of the research process with direct analogies to health care service improvement and leadership. Patients were often restricted to offering accounts of their experiences and of a “tokenistic attitude” from professionals. The review suggests that training in research methods and involvement in the whole process from design to reporting could increase patient empowerment to take a partnership role. This has clear echoes for the need to create patient leaders especially at the planning stage where they may have to most influence rather than being seen merely as sources of evaluation data.

Mockford et al. (2012) conducted a similar review directly focussing on patient involvement in the delivery of NHS services and discovered a wide variety of examples of patient involvement from condition-specific groups to board-level positions. Again the potential to exert positive influence was broad and significant but measures of impacts were lacking. There was an acknowledgement of the importance of the characteristics of individuals but there was no consideration of the training, support and development needs of patients despite their often occupying positions which would have attracted the label “leader” had they been filled by professional or managerial NHS staff. There was little economic data available to assess the cost-effectiveness of patient leadership in common with much conventional leadership research.

GenderA further significant determinant of the success of mentorship programmes, and much else within leadership development, is the effect of gender. Bickel finds a wealth of evidence that career choices, empowerment, self-confidence, social capital and many other factors important for exercising leadership tend to be lower for women regardless of professional group and level of authority (Bickel, 2014). This has significant implications for the success of leadership programmes. Focussing specifically on mentorship, Bickel argues that a key ability is to understand and incorporate these barriers into development efforts recognising the extra hurdles and challenges that may face women in leading relative to men. A further consideration for the success of programmes are widespread inter-professional gender imbalances. That such considerations are absent from many evaluations of leadership programmes may be a form of self-censorship which could explain their limited effects in altering organisational performance or improving safety cultures. In the empirical work of Chapman (2011) the styles of leadership and their differing impacts were “gendered” only in the sense that leadership behaviour employing emotional intelligence and responsiveness to colleagues concerns (traditionally “female” traits), whether or not they were employed by men or women, were associated with greater effectiveness and acceptability by followers in the various organisations studied. Only a “coaching” style was used by men more than women with other styles adopted equally by both. This review identified little evidence that gender is explicitly studied in much healthcare leadership research although it seems intuitively unlikely that leadership development needs of men and women may differ at least as much or even because of, gender disparities across professional groups and in seniority.

Interdisciplinary trainingA recent review of leadership in the NHS by the Kings Fund has been very influential in redirecting NHS leadership thinking away from traditional top-down paradigms and towards working across disciplinary boundaries (The King’s Fund, 2011). The authors point out that in recent years “leadership” has increasingly been favourably contrasted with both “management” and “administration”. Administration has now been further devalued semantically being seen as “performance measurement” and often seen as a burden. The review recommends that the NHS should develop leadership from a central hub of expertise and that this organisation should also “support the evaluation of [development] programmes, including the return on investment from leadership and management development” (italics not in the original). The authors also note that “anecdotally”, clinicians and managers buddying each other and training together can be very productive (p. 21) and should be greatly extended. Other authors agree and have called for a dramatic increase in interdisciplinary leadership training as well as better systems for identifying clinical professionals with the qualities needed to lead (Nichol, 2012).

Evaluations of mixing staff groups in leadership training is generally positive. Agius et al. (2015) studied a cohort of learners accessing an integrated medical leadership programme. This was a long standing programme running alongside specialist training and involving joint learning with NHS graduate management trainees (GMT). It was explicitly based on the medical leadership competency framework. The course led to the award of a Post Graduate Diploma or MSc in Health and Public Leadership. A mixture of interview and documentary data was used to assess the learners’ evaluation of the course and other outputs. Many candidates mentioned working alongside GMT students as a significant element in the success of the course. This appeared to be because they were already aware of the prevalence of divisions and negative stereotypes expressed about the relationships between management and medical staff within their previous working environments. Preparation to deal with the negative attitudes of peers was also rated highly as were increased non-medical technical knowledge (e.g. financial) and strategic thinking. Action learning sets applied to real service problems were also mentioned as being worthwhile.

Neily et al. (2010) measured the introduction of interdisciplinary training for entire multi-disciplinary surgical teams. Such team training was associated with double the reduction in mortality seen in departments without such training. Despite not carrying the label of “collective leadership”, the intervention contained much of the content and ethos of such theories albeit in a largely hierarchical environment such as operating theatres. Encouraging high-quality communication, governed by codes of conduct, encouragement to challenge both up and down the hierarchy and reinforcement of a team approach to working were all covered in the course which was delivered to existing teams via lectures, “group interaction” and videos.

Collaborative leadership i.e. that extends across professional and disciplinary boundaries has received some research attention (Careau et al. 2014). In a review of studies, Careau found that four leadership models predominated: “traditional”, transformational, clinical and “collaborative”. They suggest that existing leadership initiatives rarely address the competencies and skills required of systems in interdisciplinary leaders such as inter-professional communication, team working and

40 41

financial competency. Only 3.2% of the articles reviewed examined collaborative models of leadership although overall, the most common competencies studied team working and inter-professional communication. The majority of programmes were uni-professional with only 22% multi-professional and 10% inter-professional. Frenk also challenges the persistent “tribalism of the professions” and pronounces it unfit for dealing with the problems faced by 21st century healthcare (Frenk et al. 2010). They recommend that future leaders be taught how to create “collaborative and non-hierarchical relationships” and become “enlightened change agents” acknowledging that their ability to influence is more important than their formal authority. How such people are to be identified and recruited to unified programmes remains moot and the degree to which those chosen are intended to remain in their current positions or rise to managerial and executive positions is also unclear (Nichol, 2012).

Followership There has been considerable rethinking of the foundations of research into leadership in recent years. In particular the drift towards more collective and diffuse leadership models has been pronounced. Traditional concepts of leadership have tended to see three main elements as important: leaders, followers and goals (Drath et al. 2008). Rather than develop leadership per se, new models seek to “explain how people who share work in collectives produce direction, alignment and commitment” (Drath et al. 2008 – italics in the original). They argue that this gives a more effective basis for developing leadership whether of individuals or groups. The exercise of leadership is increasingly seen as dependent on followers and the alignment of their goals with those officially placed above them (Crossman and Crossman, 2011). They argue that the lack of followership research stems from an underlying belief that people don’t need to know “how to follow”. They argue that given that one aspect of followership is thought to be the ability to “lead upwards”, that is, to influence leaders, it is unlikely to be automatically nor widely understood by followers themselves as conventional views are that leadership is essentially a top-down phenomenon. To return to a continuing theme in recent literature, many now dispute the extent to which absolute leadership in healthcare either exists or is required to create clinical improvements or reliably high-value care.

Followership has received limited attention in academic research until very recently despite its being the ying to leadership’s yang. In a recent review Uhl-Bien and colleagues use two approaches to understanding followership (Uhl-Bien et al. 2014). First, followership is seen a function of formal hierarchies and casts followers as conventional subordinates. Followers may influence outcomes by complying or obstructing, essentially in response to leadership from above. Second, the focus of attention becomes the way that followers seek to “construct” their leader, resulting in the generation of social identities for both parties. This is a more radical view in which followers shape their leaders via “granting behaviours”, continual direct influence and even leading their leaders on occasion. A practical application of this second model may be in the area of interdisciplinary leadership training. Such initiatives need to negotiate hierarchies to achieve their desired effects. Adopting a followership model enables more subordinate professional groups to understand how they influence superordinate groups and vice versa in addition to helping individuals to understanding when to make such a shift themselves within their own professional structures.

An interesting synthesis of this work with developments in understanding aspects of followership is beginning to gain traction. Hayes et al. (2015) has characterised followership in the workplace, as a continuum from indifferent or negative responses to leadership direction and influence through “acceptance”, “trust” and “commitment”. Each step denotes greater identification with the aims of the leader in terms of the follower’s personal goals and reward expectations, their views of the integrity of the leader, their perception of alignment of the request with organisational goals and their ability to actually comply. A final stage of follower development is termed “stewardship” where a follower displays “extra-role” behaviours and have a creative and innovative attitude to their work. The authors describe how “conventional [leadership] thinking” can result in reduction in followership and lead to poor performance. These barriers include failure to invest in people, failure to identify long-term needs and the correct priorities, undertake appraisals as opposed to engaging in “coaching”, centralise power and break commitments amongst others. There is considerable overlap with conventional leadership programme content but the focus on the reaction of followers to these behaviours is novel.

Clearly these concepts are related to “clinical engagement” even if that generally denotes a more active than passive response to the responsibilities of followership. Leaders’ use of staff appraisal could help drive staff engagement and morale as well as setting clear objectives for performance (West and Dawson, 2012 and West et al. 2011). This is a leader behaviour to elicit a follower response but one which occurs at every level of the NHS as a routine activity.

When would leadership development be most effective? There is some evidence that the timing of leadership development interventions in terms of career progression and role transition may be key to its effectiveness. It has been suggested that in particular, multidisciplinary leadership should begin in early career before attitudes become too moulded by experience and acculturation (Stoller, 2014). Taking a medical perspective, Stoller argues that the scientific thinking and the narrow problem solving focus of medical training is actually poor preparation for leadership of multi-professional teams. For example, emotional intelligence is important for leadership but often lacking in medical training although this has changed in more recent years. Doubtless, emphasis on the doctor as lone decision maker and the consequent emphasis on individual clinical responsibility and liability, do not naturally lead to comfort with the fluidity of leadership styles thought to be most effective. Stoller suggests that to compound this, most medical staff are aware of their lack of preparedness at some level and this may result in a reluctance to take on leadership positions both formal and informal.

42 43

Curtis and colleagues concur with many authors in suggesting that the importance of leadership is now so widely acknowledged that it should form a key component of undergraduate nurse training as well as later on in clinical careers (Curtis et al. 2011). They examined undergraduate courses in Ireland, Scotland and the US and noted that the leadership content of courses was restricted to dealing with the transition from student to qualified status. It is likely that early nurse training in leadership has some important effects on subsequent behaviours and effectiveness. Demeh (2015) studied student nurses’ perceptions of clinical leadership and found that it tended to be learnt informally, often by emulating role models which may or may not be available or suitable to develop useful leadership skills. Nursing students found theoretical content in training was useful and was seen as more beneficial the earlier in the training programme it featured. Such theory may have a role in increasing the ability of junior nurses to make use of opportunities to emulate senior colleagues or select which behaviours to employ themselves.

Post-graduate leadership training is more widely available. A recent review of research on leader and leadership development (not restricted to healthcare) examines the academic research evaluating the impact of interventions in terms of career stage (Day et al. 2014). The report finds evidence that experience is more important than mere longevity in a leadership role. In addition, the career point at which different experiences occur is important in mediating their effects on developing leaders; the wrong experience at wrong time can be counterproductive for successful progress. Formal leadership programmes have a role in coordinating the introduction of new responsibilities and challenges but home organisations must also consider the environment they provide and how this can lead to successful maturation of leadership skills.

Gordon et al. (2015) collected data on medical trainees’ perceptions of their own leadership training needs from self-penned narrative accounts of their experiences of leadership and followership in healthcare. The stories were analysed for contextual, factual and performative elements. Twice as many stories involved the narrator as follower than as leader. Stories were classified as describing either standard leadership relations or “emergent” leadership situations. These latter examples were generally non-routine events such as emergencies. Alternative sources of leadership were recounted in many of these scenarios. In addition they were more likely to portrait the narrator as a leader in the situations compared with the standard scenarios. Emergent scenarios often involved complex care situations and leadership was often dependent on the emergent leader having superior technical knowledge or better judgement at that moment despite the prescence of nominal superiors. Standard scenarios were more likely to demonstrate inhibition of potential leadership in junior staff where they felt they lacked recognition for their contribution and less able to take on leadership roles. The clinical leadership content of medical training has been increasing for some time driven by work such as the UK Medical Leadership Competency Framework (McKimm and Swanwick, 2011). Despite this, doubts remain about the impact of such initiatives in the absence of supportive and nurturing workplaces in the NHS where such qualities can be honed.

Distributed Leadership Much recent literature on leadership discusses the idea of leadership residing at many levels and in different places within complex health care organisations (Gockel and Werth, 2010; Willcocks and Wibberley, 2015; Chreim et al. 2010; Martin et al. 2015 and West et al. 2014). A significant amount of academic attention has been given to variants of “distributed leadership variously known also as “collective”, “shared” or “migrated” leadership. Such theoretical models are widely seen as both a possible corrective to traditional “heroic” or “transitional” leadership and more suited to healthcare where executive power is far from absolute.

The quality of care in larger organisations is maintained by a complex interacting web of professional and managerial interventions and expectations, internal and external factors, guidelines and targets. Recent reports into failings of care (Mid Staffs, Morecambe Bay etc.) All contain evidence of failed or absent clinical, managerial and board leadership when such cultures break down. Care quality can evidently still deteriorate despite diffusion of authority, the influence of multiple professional bodies and external agencies that are explicitly designed to maintain clinical standards. For example, in the best performing acute trusts, distributed leadership and empowerment and leadership across professional groups is important in maintaining patient safety (Dixon-Woods et al. 2014 and McKee et al. 2010). In more isolated settings, where such oversight is weakest, there is an even greater need for empowerment of staff to enact distributed leadership to maintain standards, values and caring cultures and raise the alarm when poor care or mistreatment is found (Department of Health, 2012). In future more care will be delivered in smaller and more isolated settings such as care homes, often in the independent sector. Leadership development strategies must shift accordingly to ensure that quality of care is maintained ‘when no one is looking’.

Distributed leadership requires that decision-making be undertaken wherever in an organisation that it contributes the most towards achieving safety and improvement goals. Bohmer quotes Robert Naylor CEO of UCLH as saying that;

“ The more competent clinical leaders become, the easier it is for them to in turn devolve responsibilities down within the organisation. I believe complex organisations like hospitals… need to continually force decision-making down into the organisation as far as possible. The people who have direct interface with patients… are much more capable of making the right decisions at that level than someone coming down from the top of the organisation.”

(Bohmer 2012:19)

44 45

This neatly captures the essence of much writing on clinical leadership suggesting that ‘front-line’ staff are best placed to enact change safely and effectively because of the detailed understanding of the contribution they and their colleagues make to the treatment, safety and wellbeing of patients.

Difficulties remain, distributed leadership has been criticised as either impossible or as coercive i.e. allowing managerial dictates to be self-imposed by otherwise independent staff (Martin and Learmonth, 2012). Of wider concern is the risk that designations of leadership, even backed-up by suitable training, may be insufficient to allow staff to override entrenched hierarchies and power structures. Observational research has looked at the experiences of nursing staff with leadership responsibilities in settings where they work alongside staff with far greater conventional authority (Martin and Waring, 2013). The study examined the work-life experience of ostensibly subordinate staff using interview data from operating theatre staff rebranded as leaders after having received three days of leadership training (provided by LEO). Theoretical understanding of leadership concepts and practices was found to be high in those trained using the programme but there were significant problems for them in enacting leadership within their work places. Staff reported regularly feeling powerless and that they could rarely put into practice what they had learnt. One class of exceptions was when implementing managerial sanctioned policies. Another alternative way to achieve leadership in practice required them to act as a go-between for more conventionally powerful staff groups to achieve their goals. Providing such staff with practical solutions for situations where there is an imbalance of power and yet still a need to exert influence, could increase the impact of leadership development for subordinate staff who may need to rely on “soft” power to instigate change.

Other qualitative work has also questioned the ease with which distributed leadership functions in practice (Martin et al. 2015). A central theme was the perceived conflict between assertions and aspirations toward distributed models and the reality of existing political and policy contexts. Using interviews of staff who had received leadership training and development at three health provider organisations the results suggested three “disconnects” which commonly occur under distributed leadership. Distance, power and values were thought to remain difficult issues under distributed leadership with managers perceived as distant, leaders at all levels seeing power as lying elsewhere and perceived conflicts of values between managers (target driven and financially strict) and clinicians who see themselves as patient and service orientated. Managers tended to feel the clinicians were clinician orientated and financially naive with significant conflicts of interest.

A further sign of the move away from top-down leadership has been increased interest in “leader-member exchange theory” (LMX). LMX has received increasing research attention in the last decade but less so in healthcare settings (West et al. 2015). The central insight of LMX theory is that superior-subordinate relationships differ in many qualitative aspects and the relative effectiveness of leadership and its impact on performance, occur as a direct result of such variations (Erdogan and Bauer, 2014). LMX theories have focused on the view from subordinates and this does not always correlate with the views of seniors. This is possibly because senior staff feel social pressure to have equally good relationships with all their junior colleagues and therefore are inclined to contribute research data that supports such an impression (p.409). This divergence of view is itself interesting. Equal treatment of all employees by a supervisor is unlikely to meet the expectations of all if their actual needs and requirements differ. At an extreme, fairness could mean all staff are equally unhappy. Leadership development that recognises this complex relationship would be of benefit.

LMX theories are clearly are of relevance to leadership development strategies. LMX research seeks to understand transactions up and down hierarchies and examines the quality and contents of such relationships (Hesselgreaves and Scholarios, 2014). High quality LMX relationships develop over time and are characterised by increasing mutual trust, respect and greater delegation from senior to junior partner. Leaders can help to improve exchanges with their juniors by displaying ethical and empathetic behaviours, delegating effectively to demonstrate trust and displaying enthusiasm and other behaviours normally associated with “transformational” leadership styles (Erdogan and Bauer, 2014 :416). Of similar significance are the findings that LMX quality is positively correlated with employees’ perceptions of their organisation and their engagement with it. Again, given the aims of leadership development within the NHS, equipping staff at all levels with an understanding of the central importance of superior-subordinate relationships is likely to be a powerful tool for improving intermediate human resource indicators most likely to have an impact on the quality and safety of care. Erdogan and Bauer review a number of studies showing that high LMX performing staff work beyond their official job performance requirements and tend to assume broader responsibilities than expected, both of which clearly have benefits for wider organisations (p. 418). The more senior level the relationship occurs and the more complex the job demands, the more complex the LMX relationship becomes. High levels of trust and delegation, seen as markers of high quality LMX, can lead to higher demands and therefore strain in subordinates although it is also suggested that high quality LMX relationships can protect junior partners whilst increasing strains on the senior partner because of the need to maintain the relationship requiring emotional labour. Leadership training should promote an understanding of these complexities so that high-quality relationships are not only fostered but negative consequences are proactively reduced. Where LMX methods have been applied to healthcare they have generated findings of relevance to leadership development. Research in the NHS reveals a relationship between high LMX leadership style, perceptions of better support from line managers and reduced stress and intention to leave amongst a sample of 433 NHS nurses (Robson and Robson, 2016).

The demands of shifts towards more integrated care may make specific demands on leadership in the NHS. Best et al. (2012) have produced a realist review of the literature regarding “large system transformation” projects. The authors propose some “practical” remedies for achieving distributed leadership which are applicable to many change initiatives where a range of staff groups at differing organisational levels need to engage with the initiative and drive improvements. First they suggest that there should be an explicit effort to align high level and distributed leadership goals to avoid “cognitive dissonance” in staff who might otherwise see a conflict between these aims and their own motivations and concerns. Second, they note that the use of change agents can energise the required staff as well as relieve them of some of the burdens inherent in change processes. Third, they find evidence that pilot studies and small scale “proof of principle” exercises can build confidence amongst staff groups that the changes will be effective and allay fears. Finally, they suggest that leaders provide “assurance” that staff will not be penalised for implementing new ways of working and in the initial periods of institutional learning. Incorporation of the principles of distributed leadership within development programmes at any level can help potential leaders understand the reactions they might expect when initiating change and prepare strategies to ensure that these reactions do not impede progress.

46 47

A recent paper has discussed the shift towards broader conceptions of leadership and the theoretical background to shared and distributed leadership models (Turnbull-James, 2011). With leadership increasingly conceived of as “post heroic” they attempt a distinction between shared and distributed models. The report suggests that shared leadership refers to formal cooperation of existing leaders whilst distributed leadership is taken to mean organic and spontaneous cooperation and adoption of joint goals when these are seen as achievable (p.7). Professional bureaucracies can be readily recast as examples of distributed leadership (Dickinson et al. 2013) whose source of legitimacy emerges from the specific technical role of the members. In general, the concept of distributed leadership can be seen as working with, rather than against, the inter-professional nature of health care work with its diffusion of power based on exclusive bodies of specialist knowledge.

There are other reasons to welcome the shift from heroic leadership in healthcare and make this a central theme of future leadership work. Grint and Holt argue that the search for “perfect” leaders is doomed to fail (Grint and Holt, 2011). They suggest that mechanisms to temper counterproductive leadership behaviours (and failing, that to replace poorly performing leaders altogether) are as important as focusing on installing and developing such leaders in the first place. The authors join many others both from before but increasingly after the Francis Report, in calling for “responsible followership” (Grint and Holt, 2011:7 and Gibbons and Bryant, 2012). This should consist of peers, patients and families being prepared to speak up against those in authority. Leadership development needs to foster resilience and an ability to listen but followership also comes with responsibilities to ensure that standards are upheld. Leaders must be helped to understand their follower’s motivations and concerns and how they can harness these as a resource. How and when this is done may frequently on the problem being tackled. Grint suggests that healthcare problems can be “wicked, critical or tame”. Wicked problems are “complexly complex” and can appear intractable; critical problems are simpler but contain existential and urgent threats; tame problems require consistency and long-term approaches. These differing situations require “leadership, commanders or management” respectively who can supply “questions, answers or processes”. Leaders must understand which situations require which approaches as they will be required to inhabit all three roles at different times and initiate flexible and effective interactions with their followers.

The followership theme was explored in an early narrative review (Kellerman, 2007). At that time Kellerman noted that there was little literature regarding followers or followership compared to the wealth of writing on leadership itself. Increasing complexity and technical nature of the work of followers meant that expertise can frequently “trump position as an indicator of who is really leading and who is really following”. This especially true of the modern health sector. In response, leadership development interventions must pay attention to the importance of followers when introducing change and improving quality. She argues that from a leader’s perspective, followers are best categorised along a continuum of “engagement” from none to “passionately committed and deeply involved”. The proposed typology consists of five categories: isolates, bystanders, participants, activists and diehards. For example, isolates and bystanders both support the status quo within organisations but can drag performance down. Rather than suggest that leaders determine such attitudes, Kellerman suggests that that follower’s degree of engagement is largely responsible for determining the “nature of the superior-subordinate relationship”. Leadership development interventions need to equip staff at whatever level with the skills to understand this alternative framing of their workplace relationships and how they can best influence and be influenced for the good of their organisations.

Leadership needs to focus on understanding the reasons for the lack of engagement and the degree to which it should and can be increased. By contrast, activists and diehards have a quite different relationship with their superiors and time and effort on improving such relationships can have more pronounced effects given that the commitment of such workers can be for or against leaders and/or organisations and they may use their energies and commitment to either support of undermine as they see fit.

Followership remains an important consideration at higher levels within the NHS when attempting broad or transformational clinical improvement projects as these situations can allow significant alternative power structures to emerge sometimes with significant negative consequences (Dickinson et al. 2013). Such situations can present exiting staff with challenges to their traditional roles and practices and can promote a die-hard attitude against innovation. Martin and Learmonth suggest that the spread of leadership as a term, has sometimes been intended to make “policy intentions […] not just everyone’s responsibility, but part of everyone’s sense of self”. As they note, this aspiration, if held at all, is often unsupported by the findings from academic research concerning the actual experience of leaders. The reality of working life for many leaders does not appear to enable them to exert greater authority, especially against their professional instincts and many struggle to achieve change at all (e.g. Martin and Waring, 2013; Martin and Learmonth, 2012; Martin et al. 2015; Fealy et al. 2011; Edmonstone, 2014; Dickinson et al. 2013 and Davies et al. 2007).

Leadership training has a role in preparing those who are being developed for the difficulties they may face in enacting leadership roles in the circumstances they will likely encounter in the working lives. Elements of leadership training which could help overcome these barriers include Action Learning Sets, inclusion of theoretical models regarding empowerment, negotiation and an understanding of the importance of “human factors” in mediating hierarchies. The provision of high-quality longitudinal mentoring to provide greater support and the creation of an organisational culture suitable to allow leadership to flourish, are also key requisites. A lack of evidence of the effectiveness of such strategies remains for nursing for example and further research is required to understand where interventions to empower leaders in specific professions or settings could have the greatest impact (Richardson and Storr, 2010).

Adopting a realist approach to leader-follower relations and placing the emphasis on transactions in context can make sense of complex and contested situations in a way that static shared or distributed leadership models cannot always capture. By thinking of everyday situations as involving individuals in relationships where they move more or less fluidly between leader and follower roles, is intuitively appealing and underpins a lot of new thinking about leadership development. Can interdisciplinary or indeed trans-hierarchical leadership skills be actively developed? Negandhi and colleagues report on a training model developed with in the Indian Public Health system (Negandhi et al. 2015). They identified key inter-disciplinary leadership competencies as follows:

48 49

n self-awareness

n vision

n self-regulation

n motivation

n decisiveness

n integrity

n interpersonal communication skills

n strategic planning

n team building

n innovation

n being an effective change agent

Cleary this list shares many commonalities with generic leadership development content. It has been suggested that these similarities arise from the repackaging of the “old problem” of vague and poorly understood leadership theories (McCallin, 2003). McCallin proposes that interdisciplinary leadership and teamwork require an overarching “philosophy of stewardship” which is defined as a willingness to assume power without centralising it and without resorting to rewards or punishments to achieve goals. Shared leadership in interdisciplinary teams should constantly recast itself to suit the problem-space and evolving circumstances. Authority must informally shift to account for variations in the required expertise but expertise will not always be clinical. A lack of expertise in non-clinical problem-solving is widely perceived as a barrier amongst physicians for example (Bohmer, 2013). Leadership training can help clinicians see the need for and develop these alternative sources of legitimacy in their own work lives.

Leadership is a very context dependent and complex issue to unpick. This review has identified several key areas that must be addressed if leadership training is to reflect this situation and ensure that leaders of the future can operate together in such complexity.

n Practical applications of recent theories of distributed leadership are needed

n Self-awareness and appreciation own and others’ personal traits and leadership abilities across the multi-disciplinary team

n Understanding the role of leadership across systems and care pathways, including understanding of cross-boundary team functioning

n Acceptance and promotion of the importance of patient perspectives and the development of patient leadership

n Understanding the effects of followership is required in complex and highly technical healthcare organisations

n Leadership development initiatives should be evaluated using process or intermediate measures such as 360o feedback and turnover intentions rather than “high-level” patient or organisational outcomes

n Leadership development should precede or coincide with significant career transitions into roles via succession planning

n Leadership development should be targeted at isolated but key staff such as care home managers, community team leaders and general practitioners

n Standards for leadership should be incorporated into commissioning and evaluation as they already are in provider inspection regimes.

n High-level leadership shapes the culture - distributed leadership takes the culture to the bedside

Summary of recommendations from the literature

50 51

The current trajectory of the NHS must be radically altered in the coming decade to cope with unprecedented demands, without resorting to some form of rationing. Recent policy and research work on leadership in the NHS is remarkably consistent on the future direction that leadership strategy should take if not on the actual tactics. Leadership training will be required at “every level” as increased and more widespread staff engagement are essential for service improvement and provide the only route to significant performance improvements. Recent years have seen a preponderance of “pace-setting” top-down leadership within the NHS focused too intently on targets and directive management and too little on the vast numbers of “little platoons” that put quality improvement projects into effect. Greater interdisciplinary leadership development work will be required including access to external mentors and specialists where necessary. The emerging evidence from followership research suggests ways in which NHS organisation can understand and nurture strong followership skills to complement the efforts of identified leaders. Meeting narrow performance goals and requiring top-down compliance whilst sometimes necessary, can only take an organisation so far.

The current requirement to build integrated models of care across traditional organisational and professional boundaries requires leadership development with a similar ethos. Systems leadership requires more emphasis on relationship building and trust given the frequent absence of formal authority and traditional levers. Systems leaders will be, by definition, a multi-professional groups with an inter-professional mind set. There may be a need to lose the “self” of leadership altogether and work with others to improve care in ways for which no personal credit can be gained and for which leaders may not even have any official responsibility.

The idea of leadership as a matter for a relatively small number if highly influential individuals must give way to one of widespread, shifting, responsive and interacting patterns of authority exerted at whatever level it is needed to achieve an effect. For such a system to function effectively, all staff with influence on patient care at any level, must be helped to attain an understanding of how they can contribute to meeting the clinical needs and improved wellbeing of their patients. For those in senior roles this may be through a disorientating but ultimately engaging adoption of both leader and follower roles dependent increasingly on their skills and potential contributions rather than their rank or job title. For the far larger group of NHS and non-NHS healthcare staff actually delivering day-to-day care, leadership development must be accessible (online, in the workplace, delivered by immediate superiors) and demonstrate a focus on quality and safety of care that aligns with their own professional standards and priorities. Beside them will be patient and public leaders, highly skilled and informed helping to set priorities for themselves and the organisations who serve them.

To ensure that services become integrated, to reduce unnecessary variations in processes and outcomes, and to deliver services as users want to experience them, future NHS leadership must engage the skills and commitment of all paid employees, external collaborators and users. This will require a deeper understanding of the complex interplay of myriad sources of authority; a patient; a surgeon; a care home manager; and ways of increasing their ability to act as both effective leaders and followers. Developing leadership potential in such diverse groups has not been attempted in a systematic way before.

Aarons, G., Ehrhart, M. & Farahnak, L. (2014) The implementation leadership scale (ILS): development of a brief measure of unit level implementation leadership. Implementation Science, 9, 45-45.

Aarons, G., Ehrhart, M., Farahnak, L. & Hurlburt, M. (2015) Leadership and organizational change for implementation (LOCI): a randomized mixed method pilot study of a leadership and organization development intervention for evidence-based practice implementation. Implementation Science, 10(11).

Agius, S., Brockbank, A., Baron, R., Farook, S. & Hayden, J. (2015) The impact of an integrated medical leadership programme. Journal of Health Organization and Management, 29(1), 39-54.

Avgar, A.C., Litwin, A.S. & Pronovost, P. (2012) Drivers and barriers in health IT adoption: a proposed framework. App Clin Inform, 3, 488-500.

Avolio, B. & Gardner, W. (2005) Authentic leadership development: Getting to the root of positive forms of leadership. The Leadership Quarterly, 16(3), 315–338.

Avolio, B.J., Avey, J.B. & Quisenberry, D. (2010) Estimating return on leadership development investment. The Leadership Quarterly, 21(4), 633-644.

Avolio, B.J., Reichard, R.J., Hannah, S.T., Walumbwa, F.O. & Chan, A. (2009) A meta-analytic review of leadership impact research: Experimental and quasi-experimental studies. The Leadership Quarterly, 20(5), 764-784.

Baker, G. (2011) The Roles of Leaders in High Performing Healthcare Systems In Commission on Leadership in the NHS. The Kings Fund, London.

Best, A., Greenhalgh, T., Lewis, S., Saul, J., Carroll, S. & Bitz, J. (2012) Large-System Transformation in Health Care: A Realist Review. Milbank Quarterly, 90(3), 421-456.

Bickel, J. (2014) How Men Can Excel as Mentors of Women. Academic Medicine, 89(8), 1100-1102.

Biggs, A., Brough, P. & Barbour, J. (2014) Enhancing Work-Related Attitudes and Work Engagement: A Quasi-Experimental Study of the Impact of an Organizational Intervention. International Journal of Stress Management, 21(1), 43-68.

Blumenthal, D., Bernard, K., Bohnen, J. & Bohmer, R. (2012) Addressing the Leadership Gap in Medicine: Residents’ Need for Systematic Leadership Development Training. Academic Medicine, 87(4), 513-522.

Bohmer, R. (2012) The instrumental value of medical leadership: Engaging doctors in improving services. The Kings Fund, London.

Bohmer, R. (2013) Leading Clinicians and Clinicians Leading. New England Journal of Medicine, 368(16), 1468-1470.

Conclusion: What will the future of leadership development look like?

References

52 53

Brett, J., Staniszewska, S., Mockford, C., Herron-Marx, S., Hughes, J., Tysall, C. & Suleman, R. (2014) Mapping the impact of patient and public involvement on health and social care research: a systematic review. Health Expectations 17, 637–650.

Brody, A., Barnes, K., Ruble, C. & Sakowski, J. (2012) Evidence-Based Practice Councils: Potential Path to Staff Nurse Empowerment and Leadership Growth. Journal of Nursing Administration, 42(1), 28-33.

Cameron, S., Harbison, J., Lambert, V. & Dickson, C. (2012) Exploring leadership in community nursing teams. Journal of Advanced Nursing, 68(7), 1469-1481.

Care Quality Commission (2014a) How we will work together to assess how well led organsations are. Care Quality Commission, Newcastle-upon-Tyne. (http://www.cqc.org.uk/sites/default/files/Well-led%20framework%20statement%20of%20intent%20FINAL.pdf last accessed 22/01/2016)

Care Quality Commission (2014b) Working together to assess leadership. The Care Quality Commission, Newcastle-upon-Tyne. (http://www.cqc.org.uk/content/working-together-assess-leadership last accessed 25/04/2016)

Care Quality Commission (2015) The state of health care and adult social care in England 2014/15. The Care Quality Commission, Newcastle-upon-Tyne.

Care Quality Commission (2016) The five key questions we ask. (http://www.cqc.org.uk/content/five-key-questions-we-ask last accessed 13/03/2016)

Careau, E, Biba, G, Brander, R, Van Dijk, J, Verma, S, Paterson, M. & Tassone, M. (2014) Health leadership education programs, best practices, and impact on learners’ knowledge, skills, attitudes, and behaviours and system change: a literature review. Journal of Healthcare Leadership 6, 39-50.

Casey, M., McNamara, M., Fealy, G. & Geraghty, R. (2011) Nurses’ and midwives’clinical leadership development needs: A mixed methods study. Journal of Advanced Nursing, 67(7), 1502-1513.

Chapman, A., Johnson, D. & Kilner, K. (2014) Leadership styles used by senior medical leaders: Patterns, influences and implications for leadership development. Leadership in Health Services, 27(4), 283-298.

Chreim, S., Williams, B., Janz, L. & Dastmalchian, A. (2010) Change agency in a primary health care context: The case of distributed leadership. Health Care Management Review, April/June, 35(2), 187-199.

Corrigan, P, Lickey, S, Campion, J. & Rashid, F. (2000) Mental Health Team Leadership and Consumers’Satisfaction and Quality of Life. Psychiatric Services, 51(6), 781-785.

Coulter, A. (2012) Leadership for public engagement. The Kings Fund, London.

Crossman, B. & Crossman, J. (2011) Conceptualising followership – a review of the literature. Leadership, 7(4), 481-497.

Cummings, G., Lee, H., MacGregor, T., Davey, M., Wong, C., Paul, L. & Stafford, E. (2008) Factors contributing to nursing leadership: a systematic review. The Journal of Health Services Research and Policy, 13(4), 240-248.

Curtis, E, de Vries, J. & Sheerin, F. (2011) Developing leadership in nursing: exploring core factors. British Journal of Nursing, 20(5), 306-309.

Darzi (2008) High quality care for all: NHS next stage review final report Department for Health, TSO, (https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/228836/7432.pdf)

Davies, H. Mannion, R. Jacobs, R, Powell, A. & Marshall, M. (2007) Exploring the Relationship between Senior Management Team Culture and Hospital Performance. Medical Care Research and Review, 64(1), 46-65.

Day, D. Fleenor, J. Atwater, L. Sturm, R. & McKee, R. (2014) Advances in leader and leadership development: A review of 25 years of research and theory. The Leadership Quarterly, 25(1), 63-82.

Demeh, W. & Rosengren, K. (2015) The visualisation of clinical leadership in the content of nursing education—A qualitative study of nursing students’ experiences. Nurse Education Today, 35(7), 888-893.

Department of Health (2012) Transforming care: A national response to Winterbourne View Hospital. Department of Health, TSO. (https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/213215/final-report.pdf)

Dickinson, H., Ham, C., Snelling, I. & Spurgeon, P. (2013) Are we there yet? Models of medical leadership and their effectiveness: An exploratory study. NIHR SDO, 08/1808/236, (http://www.netscc.ac.uk/hsdr/files/project/SDO_FR_08-1808-236_V07.pdf)

Dixon-Woods, M. Baker, R. Charles, K. Dawson, J. Jerzembek, G. Martin, G. McCarthy, I. McKee, L. Minion, J. Ozieranski, P. Willars, J. Wilkie, P. & West, M. (2014) Culture and behaviour in the English National Health Service: overview of lessons from a large multi-method study. BMJ Quality and Safety, 23(2), 106-115.

Drath, W. McCauley, C. Palus, C. Van Velsor, E. O’Connor, P. & McGuire, J. (2008) Direction, alignment, commitment: Toward a more integrative ontology of leadership. The Leadership Quarterly, 19(6), 635–653.

Duffy, K. & Carlin, A. (2014) Enhancing the clinical leadership of band 6 staff. Nursing Standard, 29(11), 40-47.

54 55

Edmonstone, J. (2014) Whither the elephant? The continuing development of clinical leadership in the UK National Health Services. International Journal of Health Planning & Management, July/September, 29(3), 280-291.

Elliott, N. Begley, C. Kleinpell, R. & Higgins, A. (2014) The development of leadership outcome-indicators evaluating the contribution of clinical specialists and advanced practitioners to health care: a secondary analysis. Journal of Advanced Nursing, 70(5), 1078-1093.

Enterkin, J. Robb, E. & McLaren, S. (2013) Clinical leadership for high-quality care: developing future ward leaders. Journal of Nursing Management, 21(2), 206-216.

Erdogan, B. & Bauer, T. (2014) Leader-Member Exchange (LMX) Theory: The Relational Approach to Leadership, in The Oxford Handbook of Leadership and Organizations, Day, D. (ed.), Oxford University Press.

Fealy, G. McNamara, M. Casey, M. Geraghty, R. Butler, M. Halligan, P. Treacy, M. & Johnson, M. (2011) Barriers to clinical leadership development: findings from a national survey. Journal of Clinical Nursing, 20(13-14), 2023-2032.

Flodgren, G. Parmelli, E. Doumit, G. Gattellari, M. O’Brien, M, Grimshaw, J. & Eccles, M. (2011) Local opinion leaders: effects on professional practice and health care outcomes. Cochrane Database of Systematic Reviews, August 10(8).

Francis, R. (2013) Report of the Mid Staffordshire NHS Foundation Trust Public Inquiry, TSO, London.

Frankel, A. (2008) What leadership styles should senior nurses develop? Nursing Times, 104(35), 23-24.

Frenk, J. Chen, L, Bhutta, Z. Cohen, J. Crisp, N. Evans, T. Fineberg, H. Garcia, P. Ke, Y. Kelley, P. Kistnasamy, B. Meleis, A. Naylor, D. Pablos-Mendez, A. Reddy, S. Scrimshaw, S. Sepulveda, J. Serwadda, D. & Zurayk, H. (2010) Health professionals for a new century: transforming education to strengthen health systems in an interdependent world. The Lancet, 376(9756), 1923-1958.

Gibbons, A. & Bryant, D. (2012) Followership: the forgotten part of doctors’ leadership. In BMJ Careers, (http://careers.bmj.com/careers/advice/view-article.html?id=20009342 last accessed 10/01/16).

Giddings, A. & Williamson, C. (2007) The leadership and management of surgical teams. The College of Surgeons of England, London.

Gockel, C. & Werth, L. (2010) Measuring and Modelling Shared Leadership. Journal of Personnel Psychology, 9(4), 172-180.

Gordon, L. Rees, C. Ker, J. & Cleland, J. (2015) Leadership and followership in the healthcare workplace: exploring medical trainees’ experiences through narrative inquiry. BMJ Open, 5(12).

Griffith, M. (2012) Effective succession planning in nursing: a review of the literature. Journal of Nursing Management, 20(7), 900-911.

Grint, K. & Holt, C. (2011) Followership in the NHS. The Kings Fund, London.

Ham, C. (2012) Leadership and engagement for improvement in the NHS: together we can. The Kings Fund, London.

Ham, C., Berwick, D. & Dixon, J. (2016) Improving quality in the NHS: A strategy for action. The Kings Fund, London.

Ham, C. & Dickinson, H. (2008) Engaging Doctors in Leadership: What We Can Learn From International Experience and Research Evidence? NHS Institute for Innovation and Improvement, London.

Hayes, L. Caldwell, C. Licona, B. & Meyer, T. (2015) Followership behaviours and barriers to wealth creation. Journal of Management Development, 34(3), 270-285.

Hesselgreaves, H. & Scholarios, D. (2014) Leader–member exchange and strain: a study of job demands and role status. Human Resource Management Journal, 24(4), 459-478.

Hewison, A. (2004) Evidence-based management in the NHS: is it possible? Journal of Health Organization and Management, 18(5), 336-348.

Ingebrigtsen, T. Georgiou, A. Clay-Williams, R. Magrabi, F. Hordern, A., Prgomet, M. Li, J. Westbrook, J. & Braithwaite, J. (2014) The impact of clinical leadership on health information technology adoption: Systematic review. International Journal of Medical Informatics, 83(6), 393-405.

Kellerman, B. (2007) What Every Leader Needs to Know About Followers. Harvard Business Review, December.

Kirkpatrick, D. (1994) Evaluating Training Programs: The Four Levels, Berrett-Koehler, San Francisco.

Li, W.-D. Arvey, R. & Song, Z. (2011) The influence of general mental ability, self-esteem and family socioeconomic status on leadership role occupancy and leader advancement: The moderating role of gender. The Leadership Quarterly, 22(3), 520-534.

Ligon, G. Hunter, S. & Mumford, M. (2008) Development of outstanding leadership: A life narrative approach. The Leadership Quarterly, 19(3), 312-334.

Martin, G. Beech, N. MacIntosh, R. & Bushfield, S. (2015) Potential challenges facing distributed leadership in health care: evidence from the UK National Health Service. Sociology of Health & Illness, 37(1), 14-29.

Martin, G. & Learmonth, M. (2012) A critical account of the rise and spread of ‘leadership: The case of UK healthcare. Social Science & Medicine, 74(3), 281-288.

56 57

Martin, G. & Waring, J. (2013) Leading from the middle: Constrained realities of clinical leadership in healthcare organizations. Health (United Kingdom), 17(4), 358-374.

Martin, J. McCormack, B. Fitzsimons, D. & Spirig, R. (2012) Evaluation of a clinical leadership programme for nurse leaders. Journal of Nursing Management, 20(1), 72-80.

McCallin, A. (2003) Interdisciplinary team leadership: a revisionist approach for an old problem? Journal of Nursing Management, 11(6), 364-370.

McKee, L. West, M., Flin, R. Grant, A. Johnston, D. Jones, M. & Yule, S. (2010) Understanding the dynamics of organisational culture change: creating safe places for patients and staff. NIHR SDO, London.

McKimm, J. & O’Sullivan, H. (2013) Personality, self-development and the compassionate leader. British Journal of Hospital Medicine, 74(6), 336-339.

McKimm, J. & Swanwick, T. (2011) Leadership development for clinicians: what are we trying to achieve? Clinical Teacher, 8(3), 181-185.

McNamara, M. Fealy, G. Casey, M. O’Connor, T. Patton, D. Doyle, L. & Quinlan, C. (2014) Mentoring, coaching and action learning: interventions in a national clinical leadership development programme. Journal of Clinical Nursing, 23(17-18), 2533-2541.

Mockford, C. Staniszewska, S. F, G. & Herron-Marx, S. (2012) The impact of patient and public involvement on UK NHS health care: a systematic review. International Journal for Quality in Health Care, 24(1), 28–38.

Morrow, E. Robert, G. & Maben, J. (2014) Exploring the nature and impact of leadership on the local implementation of The Productive Ward Releasing Time to Care™. Journal of Health Organization and Management, 28(2), 154-176.

Mulla, A. Hewison, A. & Shapiro, J. (2014) The importance of clinical leadership in service redesign: Experience in three English National Health Service Hospitals. International Journal of Leadership in Public Services, 10(3), 126-141.

Nash, S. & Govier, I. (2009) Effective team leadership: techniques that nurses can use to improve teamworking. Nursing Times, 105(19), 22-24.

National Voices (2012) Not the Francis Report: A National Voices report on how to ensure Safety and Quality. (http://www.nationalvoices.org.uk/publications/our-publications/not-francis-report last accessed 14/03/2016).

Naylor, C. Curry, N. Holder, H. Ross, S. Marshall, L. & Tait, E. (2013) Clinical Commissioning Groups: Supporting improvement in General Practice? The Kings Fund, London.

Negandhi, P. Negandhi, H. Tiwari, R. Sharma, K. Zodpey, S.P. Quazi, Z. Gaidhane, A. Jayalakshmi, N. Gijare, M. & Yeravdekar, R. (2015) Building Interdisciplinary Leadership Skills among Health Practitioners in the Twenty-First Century: An Innovative Training Model. Frontiers in Public Health, 3, 221.

Neily, J. Mills, P.D. Young-Xu, Y. & et al. (2010) Association between implementation of a medical team training program and surgical mortality. JAMA, 304(15), 1693-1700.

NHS England (2015) Delivering the Forward View: NHS planning guidance 2016/17 - 2020/21. NHS England, (https://www.england.nhs.uk/wp-content/uploads/2015/12/planning-guid-16-17-20-21.pdf)

Nichol, E. (2012) Improving clinical leadership and management in the NHS. Journal of Healthcare Leadership, 4, 59-69.

Palm, K. Ullström, S. Sandahl, C. & Bergman, D. (2015) Employee perceptions of managers’ leadership over time. Leadership in Health Services, 28(4), 266-280.

Pawson, R. Greenhalgh, T. Harvey, G. & Walshe, K. (2005) Realist review – a new method of systematic review designed for complex policy interventions. Journal of Health Services Research & Policy, 10(Suppl 1), 21-34.

Powell, M. Dawson, J. Topakas, A. Durose, J. & Fewtrell, C. (2014) Staff satisfaction and organisational performance: evidence from a longitudinal secondary analysis of the NHS staff survey and outcome data. In NIHR Journals Library, Health Service and Delivery Research, London.

Reichenpfader, U. Carlfjord, S. & Nilsen, P. (2015) Leadership in evidence-based practice: a systematic review. Leadership in Health Services, 28(4), 298-316.

Richardson, A. & Storr, J. (2010) Patient safety: a literature review on the impact of nursing empowerment, leadership and collaboration. International Nursing Review, 57(1), 12-21.

Robson, A. & Robson, F. (2016) Investigation of nurses’intention to leave: a study of a sample of UK nurses. Journal of Health Organization and Management, 30(1).

Rosenman, E. Ilgen, J. Shandro, J. Harper, A. & Fernandez, R. (2015) A Systematic Review of Tools Used to Assess Team Leadership in Health Care Action Teams. Academic Medicine, 90(10), 1408-1422.

Rosenman, E. Shandro, J. Ilgen, J. Harper, A. & Fernandez, R. (2014) Leadership Training in Health Care Action Teams: A Systematic Review. Academic Medicine, 89(9), 1295-1306.

Santry, C. (2011) “Resilient NHS managers lack required leadership skills, DH research says”. Health Service Journal, 6 July.

58 59

Senge, P. Hamilton, H. & Kania, J. (2015) The dawn of system leadership. In Stanford Social Innovation Review, Leland Stanford Jr. University.

Shah, P., Cross, V. & Sii, F. (2013) Sailing a Safe Ship: Improving Patient Safety by Enhancing the Leadership Skills of New Consultant Specialist Surgeons. Journal of Continuing Education in the Health Professions, Summer, 33(3), 190-200.

Shipton, H. Armstrong, C. West, M. & Dawson, J. (2008) The impact of leadership and quality climate on hospital performance. International Journal for Quality in Health Care, 20(6), 439-445.

Shirey, M. (2009) Authentic Leadership, Organizational Culture, and Healthy Work Environments. Critical Care Nursing Quarterly, 32(3), 189-198.

Smith, J. Holder, H. Edwards, N. Maybin, J. Parker, H. Rosen, R. & Walsh, N. (2013) Securing the future of general practice: new models of primary care. The Health Foundation, (http://www.nuffieldtrust.org.uk/node/2764).

Solansky, S.T. (2010) The evaluation of two key leadership development program components: Leadership skills assessment and leadership mentoring. The Leadership Quarterly, 21(4), 675-681.

Sparrowe, R. (2005) Authentic leadership and the narrative self. The Leadership Quarterly, 16(3), 419-439.

Spilsbury, K. Hanratty, B. & McCaughan, D. (2015) Supporting Nursing In Care Homes. Royal College of Nursing Foundation, University of York.

Staffileno, B. & Carlson, E. (2010) Providing direct care nurses research and evidence-based practice information: an essential component of nursing leadership. Journal of Nursing Management, 18(1), 84-89.

Starck, P. & Rooney, L. (2015) Leadership for the Integration of Comprehensive Care and Inter-professional Collaboration. Clinical Scholars Review: The Journal of Doctoral Nursing Practice, 8(1), 43-48.

Stephenson, J. (2009) Doctors in management. BMJ, 339:b4595.

Stoller, J. (2014) Help wanted: developing clinician leaders. Perspectives on Medical Education, 3(3), 233-237.

Storey, J. & Holti, R. (2012) The contribution of clinical leadership to service redesign: a naturalistic inquiry. Health Services Management Research, 25(3), 144-151.

Storey, J. & Holti, R. (2013a) Possibilities and pitfalls for clinical leadership in improving service quality, innovation and productivity. NHS Service, Delivery and Organisation Programme, London, (http://www.netscc.ac.uk/hsdr/files/project/SDO_FR_09-1001-22_V05.pdf).

Storey, J. & Holti, R. (2013b) Towards a new model of leadership for the NHS. The National Leadership Academy, The Open University Business School.

Straus, S. Soobiah, C. & Levinson, W. (2013) The Impact of Leadership Training Programs on Physicians in Academic Medical Centers: A Systematic Review. Academic Medicine, 88(5), 710-723.

Swage, T. (2014) Evaluation of the patient leadership development programme pilot for Thames Valley – NHS England, Oxford Academic Health Science Network And Thames Valley And Wessex Leadership Academy. Oxford Academic Health Sciences Network, (http://tvscn.nhs.uk/wp-content/uploads/2014/05/Evaluation-of-patient-leadership-project-report-080914docx.pdf).

The Hay Group (2006) Nurse Leadership: Being Nice is Not Enough. The Hay Group, (http://www.haygroup.com/uk/downloads/details.aspx?id=2521).

The King’s Fund (2011) The future of leadership and management in the NHS: No more heroes. The King’s Fund, London.

The Kings Fund (2012a) Leadership and engagement for improvement in the NHS: Together we can. In The Kings Fund Leadership Review, The Kings Fund, London.

The Kings Fund (2012b) Patient-centred leadership: rediscovering our purpose. The King’s Fund, London.

The Kings Fund (2013) Experience-based co-design toolkit. The King’s Fund, London, (http://www.kingsfund.org.uk/projects/ebcd last accessed 25/04/2016).

Timmins, N. (2015) The practice of systems leadership: Being comfortable with chaos. The King’s Fund, London.

Titzer, J. Phillips, T. Tooley, S. Hall, N. & Shirey, M. (2013) Nurse manager succession planning: synthesis of the evidence. Journal of Nursing Management, 21(7), 971-979.

Tregunno, D. Jeffs, L. Hall, L. Baker, R. Doran, D. & Bassett, S. (2009) On the Ball: Leadership for Patient Safety and Learning in Critical Care. Journal of Nursing Administration, 39(7/8), 334-339.

Turnbull-James, K. (2011) Leadership in context: Lessons from new leadership theory and current leadership development practice. In Commission on leadership and Management in the NHS, The King’s Fund, London.

Uhl-Bien, M. Riggio, R. Lowe, K. & Carsten, M. (2014) Followership theory: A review and research agenda. The Leadership Quarterly, 25(1), 83-104.

Wallis, A. & Kennedy, K. (2013) Leadership training to improve nurse retention. Journal of Nursing Management, 21(4), 624-632.

60 61

West, M. Armit, K. Loewenthal, L. Eckert, R. West, T. & Lee, A. (2015) Leadership and Leadership Development in Health Care: The Evidence Base. The Kings Fund, London.

West, M. & Dawson, J. (2012) Employee engagement and NHS performance. The King’s Fund, London (http://www.kingsfund.org.uk/sites/files/kf/employee-engagement-nhs-performance-west-dawson-leadership-review2012-paper.pdf).

West, M. Dawson, J. Admasachew, L. & Topakas, A. (2011) NHS staff management and health service quality: Results from the NHS Staff Survey and related data. Department of Health, London.

West, M. Eckert, R. Steward, K. & Pasmore, B. (2014) Developing collective leadership for health care. The Kings Find with the Centre for Creative Leadership, (http://www.kingsfund.org.uk/publications/developing-collective-leadership-health-care).

Willcocks, S. & Wibberley, G. (2015) Exploring a shared leadership perspective for NHS doctors. Leadership in Health Services, 28(4), 345-355.

Wong, C. & Giallonardo, L. (2013) Authentic leadership and nurse-assessed adverse patient outcomes. Journal of Nursing Management, 21(5), 740–52.

Wong, C. & Laschinger, H. (2013) Authentic leadership, performance, and job satisfaction: the mediating role of empowerment. Journal of Advanced Nursing, 69(4), 947–59.

Wong, C.A. Cummings, G.G. & Ducharme, L. (2013) The relationship between nursing leadership and patient outcomes: a systematic review (update). Journal of Nursing Management, 21(5), 709-724.

Young, S. Nixon, E. Hinge, D. McFadyen, J. Wright, V. Lambert, P. Pilkington, C. & Newsome, C. (2010) Action learning: a tool for the development of strategic skills for Nurse Consultants? Journal of Nursing Management, 18(1), 105-110.

Yukl, S. (2013) Leadership in Organizations (8th ed.), Pearson Education Limited, England.

About the authorsPhilip has a background in Emergency Nursing and has studied Philosophy (BA Hons), Research Methodology (MA) and the epidemiology of road traffic injuries (PhD by Research). He manages a team of research nurses conducting clinical trials and other research in Emergency and Critical Care. He is a Knowledge and Evaluation Researcher with the East Midlands Academic Health Science Network at the University of Nottingham producing evidence reviews for a variety of local and national organisations.

Lucy has extensive experience working both within the NHS and Universities. Her clinical background was adult and district nursing and her research interests are the implementation and evaluation of innovation within organisations. Her doctorate explored the introduction of the Productive Ward into a large teaching hospital and she is currently looking at the translation of service redesign across multiple organisations. She is a Senior (Interim) Research Fellow at the Nottingham University Business School where her principle role is the management of the Putting into Practice team and the development of the enabling offer at the East Midlands Academic Health Science Network.

62 63

NotesNotes

18th March 2016