Role as both a resource and restraint
In this study, we were interested in exploring hybrid managers’ use of influence strategies and power, trying to differentiate between influence strategies used upwards and downwards in the organization. Our data illustrate how a professional background may both be a door opener and a restraint for action in both directions in the hierarchy. Callero
 has argued that when roles serve as resources, behavior may be limited and constrained because one is being denied access to other roles. Our findings are in line with this observation. Not having a medical background, nurses believed that their impact upwards in the organization was not as strong as that of doctors, and they found other ways of accessing expert power. Nurses could draw indirectly on expert power by “disguising” themselves as doctors, or by using doctors as their agents to gain strategic leverage. A different pattern emerged in the influence strategies employed downwards in the organizations. As pointed out by Currie
, because of the medical hegemony in decision-making, nurses’ influence over doctors is significantly reduced. We found that managers with a nursing background were able to draw on other types of power to achieve influence downwards in the organization. Nurses tried to be perceived as facilitators, by taking on administrative chores, thus shifting towards a referent power base.
While nurses were mostly restrained from acting on an expert base, a recurrent theme from interviews and observations of doctors was that they could not act without drawing on expert power. This was especially evident in the way that they sought to influence professional colleagues, which coincides with the expectations doctors have of professionals in management positions as the best among equals
. There appears to be a belief that simply having a medical background is insufficient for influencing medical colleagues. While a doctor might use expert power upwards in the hierarchy by virtue of being a doctor, in the same way as nurses might use a doctor as their agent, doctors believe that they have to maintain their clinical skills in order to retain credibility among peers e.g.
[41, 42]. Expert power is thus not earned once and for all, but had to be continuously reestablished and negotiated, which may represent a dilemma for doctors. For example, if a doctor relied on position or referent power in managing clinical staff, the doctors’ access to the expert base could be weakened over time as her or his status as an expert dwindled. Our findings suggest that roles do not serve to restrict behavior only because they constrain access to other roles (e.g. nurses being denied the role of doctor), but also because of the inherent expectations towards the role holder (e.g. doctors in leadership positions being perceived as the best among equals).
Roles, power and influence in a hospital setting
Our results reflect the authoritative coordination mechanisms found in hospital settings, and how managers within this setting are influenced by those mechanisms. Although our participants had some freedom in choosing influence strategies, the strategies seemed to be determined by the power bases they could access. More specifically, the emphasis that the participants placed on expert knowledge limited the influence strategies that were available. While some power bases, such as expert knowledge, are not exclusive to healthcare organizations (they area also relevant in other professional bureaucracies, such as in universities and law and accounting firms), they reflect some of the institutionalized rules and norms that exist in a healthcare context, i.e. that power lies in expertise
. Legitimate power, understood as formal authority, appeared to be less visible in the strategies used by managers in our study. Clinicians, and especially doctors, might perceive an experienced or merited doctor to have the legitimate right to influence them. This could explain why hybrid managers tend to draw on personal power, rather than position power in dealing with clinical staff
. Position power is not very effective upwards in the hierarchy either, as a manager at the department level is placed below in the formal hierarchy. Thus, not using position power reflects the separate worlds in hospitals
, where managers simultaneously inhibit the world of the formal management hierarchy and an informal, meritocracy based world. Position power seems not to be very effective in either.
Numerato and colleagues
 did a comprehensive review and argued that the dynamics and interplay between management and professionalism could be classified in five ideal outcome categories: (1) managerial hegemony; (2) co-optation; (3) negotiation; (4) strategic adaptation; and (5) professional resistance. Hybridization occurs in between hegemony and resistance, through the merging of managerial and professional skills, values, tools and knowledge. In our study, managerial hegemony and professional resistance was demonstrated through doctors sabotaging or circumventing the system in response to “unyielding” managers. Our results do not demonstrate examples of professionals taking on managerial or bureaucratic tools and logics. Adaptation was instead demonstrated by the nurse who used a doctor as an agent and the nurse who used different job titles strategically. This reflects Noordegraaf’s
 description of healthcare organizations becoming “ambiguous domains” in which expertise can no longer be isolated from other experts
Where are the horizontal strategies?
Our study suggests that clinicians might resort to using sabotage or finding informal and “illegal” workarounds. Although these influence strategies do not necessarily constitute a conscious attempt to punish top management, they have a coercive element, threatening to punish the whole organization. Further, these reactions appear more individualistic than collectivistic in nature. Indeed, a somewhat surprising finding was that we found no examples of horizontal strategies in the interviews and observations of the managers in our study. Participants appeared to be concerned mainly with their own department or professional sub-discipline, and more often spoke of other departments or hospitals in terms of “competitors” rather than “collaborators”.
 argued that coalition building, in the sense of gathering influential people together, plays a vital part in building power and influence. Ganz
 tells the story of grape workers’ ability to mobilize support from other communities through building horizontal coalitions. A similar influence strategy in a hospital setting would be to mobilize support from peer department managers, but this strategy was not present in our data. One explanation may be that that coalition building fails when managers are too focused on their own functional silos
. It should be noted that a number of Norwegian hospitals have organized doctors and nurses in separate units following the implementation of unitary management, so that managers at the lower levels of the organization only manage their own professional group. For example, a study of Norwegian health trusts in 2009 revealed that 60% of all hospitals had separated the bed units as independent units with their own management
 underscores that clinicians are trained to think on a micro-level, with clinical leaders having a micro-view focus on patients and patient service. In a sense, professionals become competitors and representatives for their own professional unit.
Various authors have pointed out that policy makers fail to understand the social structures that exist in professionalized contexts
[6, 7, 49, 50]. The results of our study could inform policy making in this area. Our study highlights some of the institutionalized rules and norms that exist in hospitals, namely the perception that power lies in expertise and that managers with a clinical background are more likely to draw on expert power than on formal position power. While nurses are restricted from directly accessing expert power, doctors are in a sense also restricted - not from accessing expert power, but from avoiding to do so - because of the importance they place on being perceived as professional role models. Decision makers and top managers need to acknowledge the social structure in hospitals and the challenges facing managers with different backgrounds, before implementing new management models and responsibilities. Our study suggests that professional roles and influence strategies should be a theme in leadership development programs for health professionals.
Methodological considerations and further research
Witman and colleagues
 point to a systematic bias in the literature on managers in healthcare, in that most of the research is based on interviews, with little emphasis on the use of observations. By using observations, a researcher can generate a partially independent view of the experiences that respondents draw on to construct their realities
. The fact that we were able to observe participants throughout their work day gave us an opportunity to produce a greater pool of data and to observe possible discrepancies between what our informants said and did. Observational data confirmed and provided additional examples of themes that emerged from interviews. Another strength of our study is that we explored both doctors and nurses’ views and experiences in the same organizations. A limitation of our study is the high proportion of male doctors and female nurses. It would be ideal to have more variation in terms of gender and professional background. We asked participants about their perception of the role of gender in relation to management and power, and they did not perceive it to be important. We believe that our results are transferable outside of the Norwegian context, as professional hegemonies are common in hospitals and other health care organizations
 and access to power is therefore likely to follow from one’s professional background, regardless of national context. We have also answered Baker and Faulkner’s
 request for the utility of the theory to be explored by applying it to more complex organizations. We applied the theory to a professionalized context and developed it further by combining it with literature on hybrid managers and power.
Future studies could investigate our findings further, for example by addressing the access to and use of power bases by general managers in health care organizations. It would also be interesting to investigate the conditions under which horizontal strategies are more and less likely to be used. Lastly, we found examples of managers circumventing and sabotaging the system. Although we deemed it beyond the scope of our paper to discuss these findings in more detail, we encourage other authors to take on a more comprehensive study of these phenomena in hospitals. Possible research questions include in what ways the formal organization of hospitals promote the use of these strategies, and whether hospitals (and other health care organizations) could be organized so that strategies which are useful for the individual are also useful for the organization.