Introduction

Teams and teamwork

The importance of teamwork and collaborative practice in health care delivery is under increasing debate (Baker et al. 2006; Wagner 2004). Discussions of adverse events in the healthcare quality and safety patient care literature suggest that the majority of causes of error are in deficiencies in non-technical skills, including communication failure, poor teamworking, poor leadership or poor decision-making (Gawande et al. 2003; Mallory et al. 2003; Runciman et al. 1993). Conversely, the smooth performance of teams has been linked to improved patient outcomes (Grumbach and Bodenheimer 2004). Thus, there are increasing calls from employers, funders and governments for institutions charged with the education of health care professionals to address teamworking, leadership and interprofessional practice as a part of core curricula (e.g. Workforce Taskforce 2007; World Health Organisation 2010).

In turn, education institutions that educate healthcare professionals increasingly recognise the development of teamworking capabilities as core graduate outcomes, reflecting the demands of the standards of professional bodies that describe the capacity to work in healthcare teams as an important professional attribute. The CanMEDS framework, for example, suggests that doctors should “demonstrate leadership in a healthcare team” where this is appropriate (Royal College of Physicians and Surgeons of Canada 2005, p. 23) and the Nursing Council of New Zealand (NCNZ) expects nurses to promote the “nursing perspective” in a team, collaborating with “members of the health care team to facilitate and co-ordinate care” (NCNZ 2007, p. 18).

Such policy agendas and statements assume commonsense and ahistorical understandings of work teams and teamwork; of people with complementary talents working together (often with a level of discretion over the work) to achieve a common, shared outcome (a definition that is applied here whenever teams are referred to). However, theorists such as Engeström (2008) point out that while teams have become popular forms of organising and managing work, there is little in the literature that examines the processes of teams and teamwork in real contexts—that is in clinical practice. Rather than being delivered by defined bounded groups, the members of which habitually work together, health care is more likely to be delivered by loose coalitions of health professionals formed ‘at the bedside’ to address a patient’s healthcare needs in what Engeström (2000, 2008) refers to as knots and knotworking rather than teams and teamworking.

Some policy documents do go some way to recognising such contested ideas with respect to teams. For example, issues of leadership and concepts such as ‘shared leadership’ are increasingly enshrined in policy rhetoric (see for example NHS Institute for Innovation and Improvement and Academy of Royal Colleges 2008). However, issues such as, who might lead at any one time, of what form that leadership might take and the nature of this leadership in healthcare settings are less frequently addressed. Leadership literature reminds us also that effective leaders need followers (Densten and Gray 2001; Grint 1999). However, no-one leads all the time, and followers are very rarely passive, especially where they are professionals for whom autonomy is a desired attribute. The concepts of ‘little leaders’ (who lead in small ways) and ‘active followers’ (Kelley 1992) are increasingly described in distributed leadership models that are becoming more common in health care (see for example Department of Health 2006). Erroneous assumptions in health care delivery about who might lead and who might follow, of when they might do so and what forms such leadership and followership may take may result in overt conflict, tacit followership or subversion, leading to negative impacts on patient care.

Reeves et al. (2008) suggest that “whilst doctors and nurses are certainly not the only figures in the healthcare team, there is no doubt that the status of medicine and the sheer size of nursing continue to ensure that any successful model of service delivery relies upon the effective collaboration of these two professional groups” (2008, p. 1). This article explores the nature of leadership and followership in collaborative health care provision by doctors and nurses as members of interprofessional groups. We also consider how members of different professions (specifically, in the context of this paper, the medical and nursing professions) work together to provide health care, exploring the shift from ‘professionally anchored care’ to collaborative care (Reeves et al. 2008). It also considers the influence of identity and power relations on these concepts. We draw on a subset of data generated from a New Zealand study involving hospital-based second-year resident medical officers (junior doctors) and nurses 2 years from graduation to illustrate a discussion of these ideas and their interaction in healthcare settings. Drawing from the data descriptions of the micro-culture experienced by hospital-based junior doctors and nurses we consider what leadership means, how this relates to development of professional identity at a time of transition and how stereotypes and misunderstandings may develop. We also discuss implications for educators of health professionals in developing collaborative and leadership capabilities.

Formation of professional identity

A range of ideas about identity and how it might be formed arise from the various epistemological traditions that have informed theorising about it (Kreber 2010). However, there is consensus that identity is socially constructed, not a given, and that identities are multiple and fluid (see for example, Bleakley 2006; Kreber 2010; Wenger 1998).

New professionals are subject to a range of social experiences. They are a subject of their own background and culture; recent (and varied) educational experiences; the experienced norms and behaviours of the professional groups of which they are junior members, and society’s expectations of the nature of the work they do and the manner in which they might do it (Degeling et al. 2003). They are subject to the views of society at large which (in an oversimplification) frequently characterise the professional groupings relating to medicine and nursing as a hierarchy arising from discourse which stereotypes the doctor as the ‘male/masculine curer’ and the nurse as the ‘female/feminine carer’. Research indicates that incoming health professionals (i.e. new students) hold such assumptions about the nature and role of their chosen profession as well as that of other professions (Horsburgh et al. 2006). However, the experience of the work and activities of healthcare settings and the policies which underpin these are also highly influential. Wenger (1998) describes how novices are ‘admitted’ to a ‘Community of Practice’ (CoP) through the mechanism of legitimate peripheral participation. The novices are defined as part of the CoP as ‘apprentices’ and, through a series of rites, rituals and activities, supported by experts, masters or mentors from the community, are facilitated to become part of it, travelling on an inbound trajectory to become expert practitioners themselves (Davis 2006). Wenger (1998) refers to this as “layering of events of participation and reification by which our experience and its social interpretation inform each other” and come together to construct who we are through the “constant work of negotiating the self” (p. 151). This may result in the internalisation of certain ways of being and behaving as health care professionals, particularly as leaders and followers in the provision of collaboratively delivered health care.

Varied legal frameworks, practices and histories bestow positional power on members of professions, who achieve and maintain professional membership through participation in rituals, rites of passage and conformity to rules. Some of these are reified by being formalised and written down (such as professional standards and registration requirements), others are unspoken, informal and fluid. Although ‘the professions’ themselves act as mediating artefacts on the development of individual identity, the identity of professions is not fixed, but is, in turn mediated by external influences and internal behaviours and responses. Such a dynamic system further complicates professional identity formation for new healthcare professionals and may result in individuals being out of step with the current identity of professions with whom they engage, leading to ‘role conflict’ or dysfunctional interactions with others.

The study

This article draws on a subset of data gathered from hospital-based second-year resident medical officers and nurses 2 years from graduation. These young professionals were chosen as participants because, being at a time of transition from student to novice practitioners, they have to deal with competing images and messages and their felt impacts as they start to perform in the role of a more independent health professional (Warin et al. 2006) without the buffer of being a student sheltered within the university context.

Ethical approval for the study was obtained from the Human Participants Ethics Committee at the University of Auckland. The study involved detailed semi-structured interviews with a range of participants and a follow up questionnaire survey. Sampling was opportunistic. Any individuals who met the general requirements for the study and were willing to participate were included.

The semi-structured interviews explored issues related to professional identity and belief formation, experiences of providing health care, and of working with other healthcare professionals. These themes were developed following a review of the literature related to interprofessional practice and inteprofessional care (see for example, Zwarenstein and Bryant 2000). The interview schedule was piloted with a novice doctor and nurse and changes to order, prompts etc. were made. Notwithstanding the discussion about teams above, participants were asked to describe situations in which they had been involved in the provision of health care in interprofessional teams—situations where they were expected to work with professionals from other groups to care for patients. We asked participants to describe specific instances of such situations and, during the interview, referred back to the instances to explore the ideas they were expressing.

A saturation technique was used to determine the number of participants. When fresh interviews were not adding anything novel to the data across the areas of the interview schedule (Cousin 2009) no more interviews were conducted. This resulted in thirteen resident medical officers and twelve nurses being interviewed. Interviews were conducted by three interviewers (two of the authors, MB and SG, and a research assistant) and were carried out face-to-face or by telephone. To help maintain consistency of the interviewing approach interviewers worked in mixed pairs for the first interviews. Interviews typically lasted between 50 and 70 min. The interviews were audio-taped and transcribed and coded with the help of nVivo 8 software. Coding of the data was carried out by one author (MB) with a second author (JM) checking and agreeing on the coding. In the first instance, analysis was carried out using a coding scheme based on the main areas covered in the interviews. Second order coding was then carried out against themes recurring in the data. The second order coding is important in the context of this article as the first sweep through the data alerted the authors to the fact that the participants were raising many ideas related to leadership and followership, although these were not notions specifically targeted in the design of the interview schedule. For this reason an on-line survey was developed; particularly designed to draw out data that would both supplement and complement the leadership and followership ideas that were emerging from the analysis of the interviews. The survey was sent to all those meeting the general requirements in hospitals in the local region While the response to the survey was poor, only those survey data that show statistically significant differences (p < 0.05) are reported here.

Findings

In this article we draw on interview and survey data across three related themes. During the initial coding of the interview data we became aware that participants frequently linked ideas of collaboration and of who leads and who follows with attributes that they were ascribing to themselves and other healthcare workers, about different ways in which they and other groups appeared to view the purpose, authority for, and activities of their work. In this section, we discuss the data under three related themes relating to the development of professional identity as a leader and follower:

  • gaining authority

  • power relations and the exercise of power

  • management, leadership and ‘followership’.

Gaining authority

The data suggested that the early stage professionals (nurses and doctors) in the study gained their authority for action from sources of a very different nature and further, that each group reacted differently to situations of leadership and followership; members of each group using qualitatively different approaches when making decisions about and changes to patient care.

All the nurses we interviewed discussed the ways in which their authority arises from their control of the ward or operating theatre environment, which was derived, in particular, from their understanding and implementation of formal ward or hospital policies and by carrying out the processes and procedures associated with them. JN9 expressed this thus: “the protocols give nurses some rights to do some (things such as) the emergency response to the patient”. Similarly enforcement of ward policies enabled nurses to establish and control the culture and environment of the ward. For example, one nurse’s description summarised the general view of nurses who sought to guide a doctor who might have a different idea of how certain care should be delivered when she said (to the junior doctor) “actually this is our ward policy and this is what we need done” (JN4).

The interview data from the junior doctors consistently suggest a different source of authority which arises from their position as a diagnostician and their perceived accountability for clinical decisions. JD2 described it as—“we are probably, the overriding authority, on top, on what’s the main goal of therapy”. This is typical of a hierarchical command and control conception of medical decision-making which was evident in the views of all junior doctors who discussed these ideas. For example, JD4 stated: “Whatever I say goes until someone higher up the medical team counteracts me”.

As well as their view of themselves as diagnosticians, the junior doctors frequently expressed their resistance to certain forms of authority. This was expressed by many as a resistance to the very policies, processes and procedures from which nurses derive authority. For example JD8’s commented that “the doctor doesn’t (want to) know what particular protocol for that ward is” and a seemingly ‘naïve’ understanding of systems and processes that JD13 applied to nursing: “(nursing is) much more strictly guided in that they’ve got protocols that they need to follow and it seems much more regimented”.

Power relations and the exercise of power

The previous section suggested that junior nurses and doctors in this study derive authority from sources of a very different nature. This section considers the extent to which this influences the power relations that are in play between them. As we considered the participants’ responses to questions about their relationship with various work colleagues, the work of Foucault informed the analysis. This meant looking for data which illustrated different sorts of relationships between individuals (Foucault 1988), particularly where the actions of one party may modify or constrain the present or future actions of another (Foucault 1983) and also considering the context in which the relationships were occurring, as the how of power will affect its operation and effect (Foucault 1983).

All the participants in this study worked in hospital settings and were subject to a range of bureaucratic and human resource controls. However, as registered health professionals, these novice nurses and junior doctors often expressed a view that the most important accountability that they had was to their patient(s). This primary accountability (conferred on them by a registration authority) provided these health professionals with an opportunity to exercise power or resist its exercise in a range of ways. This accountability may cut across reporting lines because “everything we do is about the patient and our accountability really should start and finish with the patient” (JN1). On the occasions when health professionals are delivering collaborative patient care, their joint accountability to the patient provides avenues for resistance to exercise of power by others and the opportunity “that you know that they (in this case doctors) are working with you… not over, kind of thing” (JN4). Such statements indicate a looseness in structures that is commonly perceived and summed up by JD3: “The team doesn’t really have an official organisation or co-ordination, its… just an understanding that everyone involved in the patient care belongs to that group of people that are involved with the patient” or JD7’s view that responsibility for a team is “a shared responsibility”.

The ‘capillary power’ described by nurses in the previous section in their use of policy and procedures is a power relation that is infiltrating rather than dominating, working horizontally through active resistance and in subtle ways (Bleakley 2006; Foucault 1983). The way in which junior nurses conducted themselves was influenced by a resistance to a form of sovereign power which they perceived (in the context of interprofessional working) to be in the hands of their medical colleagues and which was enacted by the wielding of authority over others. Their resistance to such power, by finding ways to work around its exercise, caused them to conduct themselves in particular ways. For example, allied to their knowledge of policies and procedures, many of the nurse participants described ways in which they mobilised relatively sophisticated understandings of systems and hierarchies to indirectly affect changes to patient care if they felt that their accountability to a patient’s welfare was being undermined—e.g. “If it was to do with patient care, and there was something the doctor wanted to do, but we didn’t agree with it, we can go to our charge nurse” (JN4).

On the other hand the junior doctors in this study also appeared to rely to a far greater extent than a novice nurse was able to, on the imposition of a type of sovereign authority (McHoul and Grace 1993) associated with the vertical power relations derived from a medical hierarchy. Commonly the junior doctors saw themselves (and their senior medical colleagues) as “the main driving person in the patients’ care” (JD8) with nurses ensuring that the doctor’s directions are carried out and with doctors being aware that they are reliant on nurses to do this work.

Management, leadership and followership

The rather different views that these novice professionals hold about sources of authority and the exercise of power leads us to consider the ways in which they might then work together in interprofessional healthcare settings, specifically how notions of leadership and followership typically associated with such settings might be played out. An analysis of questionnaire data suggests divergent views in these areas also.

When asked to select the leadership attributes that belonged to doctors and nurses:

  • Both doctors (100%) and nurses (88%) saw the ability to provide leadership as a core attribute of doctors. Further analysis of the strength of feeling about this shows a significant difference in the mean values between the two groups when asked (on a 1–4 agreement/disagreement scale) about the strength of the importance of this ability as a core attribute (meanD = 3.73, meanN = 3.09).

  • 94% of nurses saw the ability to provide leadership as a core attribute of nurses—however, only 55% of doctors agreed.

  • 70% of nurses thought nurses should make decisions on behalf of an interprofessional team, but 80% of doctors disagreed.

  • 78% of nurses thought nurses should contribute to making diagnostic decisions, whereas only 50% of doctors believed this should be the case.

Additionally, while 90% of the junior doctors strongly agreed or agreed with the statement ‘Effective interprofessional teams rely on strong leadership from doctors’ only 55% of nurses did. Doctors more strongly endorsed this statement than nurses did (meanD = 3.45, meanN = 2.24, scale 1–4). In contrast only 64% of junior doctors strongly agreed or agreed with the statement ‘Effective interprofessional teams rely on strong leadership from nurses’ compared to 81% of nurses.

These data suggest that the social and societal experiences of these junior doctors result in them becoming active followers of hierarchy that makes the role of ‘doctor-as-leader’ a commonsense position for them to take. A position reinforced through being given opportunities for “little leadership” (Kelley 1992) as it “feels good to be in charge of a ward round with patients looking up to you” (JD1). Thus their place in relation to their colleagues, the ward and patients is reinforced and validated. The same may be said for the novice nurses in this study who clearly see doctors as having leadership roles in health care. So, the (even relatively junior) doctor’s role as clinical leader is validated by both professional groups and patients and reinforced through rituals (such as ward rounds) and the positional power of the medical team.

However, the survey and interview data suggested that the participants had different views when it comes to the role of the nurse with respect to leadership and followership. Nurses clearly saw themselves as having a leadership role in health care delivery whereas doctors are less convinced of this. We do not know specifically what the respondents mean by ‘leadership’, but what we are told is that whilst nurses saw themselves as decision makers within collaborative health care provision, doctors were less likely to see nurses in this role, and only 50% of doctors saw nurses as contributing to (not making) diagnostic decisions, whereas 78% of nurses saw making clinical decisions as part of their clinical role.

Discussion

The varied ways in which the young professionals who participated in this study described their collaborative working could be considered using a number of theoretical frameworks, for example communities of practice, actor-network theory, or activity theory and the work of Engeström referred to earlier. However, in this article we are particularly interested in exploring the relationship that was revealed in the data, between the participants’ views of management, leadership and followership, and the different ways in which the novice doctors and nurses in this study gain authority, the power relations that these allow and the identities which they are developing.

The different conceptions of power, authority and leadership that these novice health professionals hold arise from the social and societal influences to which they have been and are being exposed and the ways in which they react to these—that is, both structure and agency are important (Collinson 2006) when considering the complexity inherent in settings of leadership and followership. Such influences come from a range of sources—from the different educational backgrounds of these professionals, from the structure and nature of their professional roles as set out by legislation and legislative bodies such as registration councils and as a result of the nature of the organisation of hospital-based health care which places the participants in this study in positions relative to one another.

Our data suggest that both groups being considered in this study ‘aspire’ (even expect) to lead (Kouzes and Posner 2002). However, what this means in practice is relatively unformed. These junior practitioners appears to hold differing ideas of leadership and followership that are based around preconceptions, stereotypes and assumptions relating to their own role as leaders and to the role of ‘the other’ as leader or follower. Far from working within some of the ‘new’ leadership paradigms that emphasise collaborative, shared, transformational or distributed leadership (Storey 2004; McKimm and Held 2009), these junior doctors seem to be working within a more transactional frame, based around ‘command and control leadership’.

This view of decision making is in harmony with the views described in the Tooke report of the “doctor’s frequent role as head of the healthcare team and commander of considerable clinical resource…. An acknowledgement of the leadership role of medicine is increasingly evident” (Tooke 2008, p. 90). This also relates to their perceptions of power relations and authority being based around sovereign power.

Doctors also appear to see a sharp distinction between management (seen as inferior to leadership) and leadership. The doctors’ perception reflects a more traditional perception of leadership and management. As Gosling and Mintzberg note:

the separation of management from leadership is dangerous. Just as management without leadership encourages an uninspired style, which deadens activities, leadership without management encourages a disconnected style, which promotes hubris [pride, arrogance]. And we all know the destructive power of hubris in organizations… (2003, pp. 54–55).

For nurses, on the other hand, the distinction between management and leadership is more blurred; management is seen as an essential part of healthcare delivery. The nurses who participated in this study were based in wards or operating theatres and were able to control and influence its environment. The nurses’ power base is in ‘the room’, the place where she/he belongs and about which she/he has a great deal of knowledge. This makes them a part of the culture and provides opportunities for management (transactional leadership) and the application of well understood protocols and procedures.

However, the nurses’ role in running a ward or managing a department and ensuring the smooth running of the service and patient care does not seem to be seen as leadership by the junior doctors. Instead it is somewhat dismissed as simply adherence to ‘procedure and protocol’ which sometimes gets in the way of clinical activity. The nurses’ expectation that they will lead clinically and make clinical decisions is dismissed by many of the junior doctors, but also the junior nurses do not see the junior doctors as leaders in their own right. This reflects Collinson’s discussion about identity not being fixed at birth by say religion or gender, but needing to be “recursively earned and achieved” (2006, p. 182). Not only does this apply to the individuals in this study but also to the two professions as a whole where the disputes at the boundaries (for example around prescribing rights which suggests nursing taking a limited role as diagnostician—shifting from carer to curer) lead to a need for a recalibration of identity and hierarchy. Coombs and Ersser (2004) examined clinical decision-making in intensive care units. They concluded that medical knowledge prevailed while nursing knowledge (derived from continuity of care) remained supplementary to it. In a similar way, the junior doctors interviewed in this study are invested in a power/knowledge relationship that results from the ‘art’ of clinical decision-making processes which are not perceived as formally codified in the ways that nursing policy and processes are and may therefore be less amenable to resistance.

However, (and closer to the subject of this study) the nurses’ reliance on codified policies and procedures sees nurses moving into the more positivist world of the diagnostician’s application of scientific principles to the diagnostic task of the curer. Generally, the development of bureaucratic mechanisms, such as clinical guidelines and protocols, which are evidence-based and codify current accepted practice allowing care to be managed through standardised pathways, has facilitated the assumption of new roles and additional responsibilities by non-medical personnel (Colyer 2004). Such protocols enable and empower nurses to act in particular ways, while at the same time constraining them from crossing professional boundaries (whether articulated or tacit) or disturbing hierarchies. Their development is an example of a particular power/knowledge relationship where the development of a new knowledge (the scientisation of nursing knowledge resulting in written procedures) has produced new types of power for nurses. Thus, the procedures and protocols form a truth game—a set of systems and procedures which incites individuals to consider their identity in relation to the accepted norms, behaviours and attitudes of their profession (Foucault 1984)—in which nurses are invested and from which considerable authority arises. The exercise of this authority, in some instances described in this study, might be considered as an exercise of monitory democracy, with nurses coming to “feel that they can put a stop to bossing, that they are the equal of others, that they have it in themselves to change things, or to keep things the way they are” (Keane 2009, p. 709).

Junior doctors are still ‘in training. They typically work across clinical facilities and regularly move from one type of ward or other clinical setting to another, attached to different groups of consultants and their teams. This contrasts with the nurses in this study who were located in one ward or unit, working with a defined, relatively stable team and deemed to be young practitioners. Although the junior doctors have a defined status as part of the medical community of practice (which is often seen as a higher status profession than nursing), they are further back on the inbound trajectory towards becoming expert practitioners than are the nurses who have completed formal training. Junior doctors do not perceive that the sort of power and authority available to their nursing colleagues is available to them, thus they are not able to use (to the same extent) knowledge of specific protocols and procedures as a source of authority. Instead they are able to draw on a type of sovereign power (Foucault 1983), working through traditional, vertical authority-led hierarchies (Bleakley 2006) which they quickly learn to use in order to develop some authority.

Power runs through every encounter and in subtle ways. Where sovereign power works by imposing authority over others, capillary power often works through active resistance to such authority from below, with a view to making changes for the better. Capillary power is an infiltrating presence rather than a dominating force. The constant rub of these different types of power relations in the collaborative care setting leads to tension between the groups participating in this study and the differing perceptions of power and its exercise and of leadership may militate against effective collaboration and communications. As many writers have noted (see Stein 1968; Fagin and Garelick 2004) nurses use the unwritten rules of the ‘nurse-doctor game’ to influence healthcare decisions. Junior nurses in our study do not passively ‘follow’ junior doctors, but choose to follow when they believe it is in the patient’s interests to do so. Neither do junior nurses actively challenge the sovereign power of doctors and the medical team, instead they might challenge through a more senior nurse or use relationships or processes (such as waiting until after a shift change) to subvert authority.

Individuals in this study come up against the ambiguity of their (fragile and contingent) identities in the interprofessional care situation they describe. The questionnaire data shows the different expectations of each group with respect to the taking of leadership (and presumably therefore followership) roles. It is not possible for a member of either profession to “define a clear, coherent and consistent self as either subject (for example, leader) or object (for example, follower)” (Collinson 2006, p. 183). Thus their identities remain ambiguous and insecure, particularly if there is an expectation that one might move between subject positions as leader in some situations and followers in others.

The differences in the power, authority and leadership paradigms that mould the identity of doctors and nurses will always create difficulties in working together if we do not move beyond the traditional. This study focussed on junior staff at a transitional stage and we might anticipate different findings from senior established health professionals. However, perhaps these junior doctors and nurses are emerging from higher education institutions with these ideas and at this transition stage they seem unable to make the incommensurate beliefs work together.

This study also reveals some differing conceptions around the nature of the healthcare ‘team’ and the difficulties applying this term in healthcare settings might create. As service delivery becomes more fluid and integrated, health professionals typically belong to more than one ‘team’ as part of their day to day working. The need for health professionals to learn to work collaboratively in multiple teams or knots (and thus manage multiple identities and relationships) is considered essential for the delivery of effective health care (World Health Organisation 2010). These configurations include uniprofessional groups (the ‘medical team’, the ‘nursing team’) which have core and permanent membership; ad hoc membership (e.g. specialist nurses) and some transient membership (students, trainees, locums). Interprofessional teams might be permanent and linked to a geographical location or clinical context (e.g. operating theatre, clinic or general practice), but in hospitals they often comprise a transient grouping formed ‘at the bedside’ or ‘informal’ interactions of medical and nursing staff members. We might think of the latter more as a loose ‘coalition’ than a team. Traditional leadership/followership models may help understanding of the former but not the latter and other paradigms and approaches may be more helpful. For example, in our study, the junior nurses’ permanent ward member status might enable them to take charge of ‘the room’, take a leadership (‘disguised’ as management) role and use capillary power. Junior doctors, although permanent members of the medical profession, are only temporary members of the medical team and the ward team in this clinical setting. In some ways, the junior doctors do not have a ‘core’ or ‘home’ team whereas the nurses’ teams are synonymous. The transient nature of the junior doctors’ working life may lead to more difficulties in developing a robust professional clinical identity which can cope with the fluidity of working within multiple groups. This may be one explanation of the more traditional and uni-dimensional perceptions of the junior doctors, whereas the nurses with their more secure grounding are becoming more adept at playing the ‘game’ (or are choosing not to play a game at all) but also may have a relatively more sophisticated view of the reality of contemporary health systems, team relationships and processes (Stein et al. 1990).

Conclusions

In this article we have used our data to prompt consideration of the views that young professionals are bringing to their professional lives and in particular to their interactions with other health professionals in the collaborative delivery of health care.

This study has implications for practice in helping to explain the tensions arising in these situations from different ways in which power, authority and leadership are perceived and exercised. The nature and inter-relationships between professions is constantly shifting. In New Zealand, for example, the changing role of the nurse as an advanced practitioner with extended roles (such as leading and managing nurse-led clinics) may lead to additional expectations by nurses related to contributing to and leading clinical decision-making. Such changes need to be carefully managed as they may be seen as a threat to doctors’ traditional power bases with subsequent resentment, confusion and misunderstanding on both sides. Service delivery change needs to be managed in ways that enable health professionals to continue to have patient-centred conversations, articulate role boundaries and overlaps, and identify and learn different ways of working.

This study also has implications for university staff, clinical teachers and those involved with the induction of new graduates to health care. In particular it raises issues about the way in which curriculum might be developed in order to address what young professionals learn about each other and their ways of working. Reflecting an extensive body of research and practice we might do this through interprofessional learning; early positive experiences of collaborations with other health professionals in the practice context; good role modelling from different professionals; educating students to challenge stereotypes, learn about other professionals and develop respect for others.

While this learning needs to address collaborative working, foregrounding patient safety and quality health care it needs to go beyond simply addressing non-technical skills commonly associated with collaboration such as decision-making, communication and understanding the roles of other health professions (Gawande et al. 2003). The data in this study suggest that different groups of new graduates have qualitatively different conceptions of how to work together, particularly around notions of leading, following and managing. While shifting healthcare coalitions might “require constant questioning and reconfiguration of the division of labour, rules and boundaries of the team and the wider organisation” (Engeström 2008, p. 230), the junior doctors who participated in this study seem to mobilise traditional hierarchical models to guide their interactions. In contrast, nurses draw authority from policies and procedures and their more fixed physical location to resist some of the leadership that doctors aspire to provide.

We would agree with others who emphasise the need to move beyond teaching to improve communication towards a greater emphasis on negotiation strategies. For example, Engeström (2008) suggests that negotiation is central to operating successfully in knotworking where the object of the activity is “unstable, resists attempts at control and standardization, and require rapid integration of expertise from various locations and traditions” (p. 230) (certainly a characterisation apt in the healthcare setting). Similarly, in discussing the collaborative working required in operating theatre or intensive care unit settings, Lingard and her colleagues also emphasise the need to teach negotiation (for example, Lingard et al. 2004).

This study suggests that the power relations experienced by novice professionals have a productive effect on them, moulding their identities and causing them to mobilise a range of strategies as they attempt to move from peripheral to central participation in the community of practice. The development of novice professionals who can build the capabilities to negotiate the various roles of leader, follower and manager that they will play in collaborative care may be enhanced if supported by educational interventions which seek to help students and novice professionals gain a better understanding of the ‘how’ of power (Foucault 1983) in healthcare settings.

Such understanding may help turn the policy rhetoric of ‘shared’ leadership and collaborative interprofessional working from a theoretical concept to a practice reality for these junior nurses and doctors.