Professional collegiality and peer monitoring among nursing staff: An ethnographic study

Abstract

Background

Patient safety and professional self-regulation systems both rely on professional colleagues to hold each other accountable for quality of care.

Objectives

To understand how staff nurses manage variations in practices within the group, and negotiate the rules-in-use for quality of care, collegiality, and accountability.

Design/Methods

Ethnographic case study; participant-observation, semi-structured interviews, policy analysis.

Setting

In-patient unit in an urban US teaching hospital.

Results

Explicit acknowledgement of conflicts and practice variations was perceived as risky to group cohesion. The dependence of staff on mutual assistance, and the absence of a system of group practice, led to the practice of “mutual deference”, a strategy of reciprocal tolerance and non-interference that gave wide discretion to each nurse’s decisions about care.

Conclusions

Efforts to improve professional accountability will need to address material constraints and the organization of nursing work, as well as communication and leadership skills.

Introduction

Peer relationships in nursing are important for several reasons: coordinating care across shifts or sites; ensuring quality and safety; and establishing and maintaining the standards of professionalism. Collegiality – the relationships of colleagues working together – provides a framework in which professionals are expected to hold each other accountable for the care that is given, and for its congruence with local and national standards (Hansen, 1995). Collegiality is therefore the first line of professional self-regulation. Previous research reports have suggested that staff nurses (much like other professionals) tend to avoid conflict with peers, however, and particularly to avoid discussions of errors and problems (Mahon and Nicotera, 2011; Roberts, Demarco, and Griffin, 2009; Siu, Laschinger, and Finegan, 2008). Explanations for this avoidance have tended to concentrate on psychological and cultural reasons (Brinkert, 2010; Mahon and Nicotera, 2011; Valentine, 2001), with relatively less attention paid to organizational factors (Davies, 2003).

This paper describes an ethnographic study of a hospital unit in the United States, focused on how the nursing staff negotiated everyday standards of practice with each other. The results of this study suggest that the practical incentives for maintaining positive relations with co-workers, and the corresponding disincentives for risking disruption of those relations through questioning or challenging the practices of others, leads to a system of “mutual deference”, a strategy of reciprocal tolerance and non-interference that gives wide discretion to each nurse’s decisions about care. Efforts to change nursing behaviors that focus only on communication skills and becoming “comfortable with conflict” will be unlikely therefore to have much effect.

Background and Significance

Central to most notions of professionalism is the idea of self-regulation: the obligation of a profession to regulate its own membership, rather than being regulated primarily by the state (Freidson, 1970). In this model, a profession essentially offers a contract with society: in exchange for a great deal of latitude in regulating its own affairs, the profession pledges to protect and promote the interests of the public, and not merely the interests of its members. Professional regulation occurs most visibly through formal systems such as educational preparation, licensure, and nursing practice legislation. These formal systems are mainly used to handle extreme or egregious cases, however, and function as tools of last resort more than everyday guides to conduct (Raper and Hudspeth, 2008). Some researchers have therefore turned toward informal systems, such as the “unwritten rules” and cultural norms of the workplace, in order to explore professional regulation (Ehrich, 2006; van Maanen and Schein, 1979).

A crucial part of collegiality is accountability, which means most literally being required to give an account, having to answer when asked (Scott and Lyman, 1968). More broadly, accountability means that professionals are answerable to each other for their practice, in relation to professional standards. Collegial relations in nursing are consistently described as falling short of this ideal of accountability, however (Almost, 2006).

The challenges of intra-professional relationships are not unique to nursing. Fieldwork studies on peer relationships in, for example, medicine (Bosk, 1979; Cassell, 1991; Friedson, 1975) have generally found a reluctance on the part of professionals to confront, challenge, or discipline each other. This creates a paradox in the sociology of professions, in which a crucial attribute according to the theory – self-regulation – is also widely acknowledged to be rare and weak in actual professional practice (Abbott, 1988).

Nursing practice offers an important site to investigate this issue, for nurses – especially though not only in hospitals – have a comparatively high degree of intra-disciplinary collaboration. Nurses often work in close physical proximity, share tasks and information, trade patients back and forth across shifts, seek each other out for problem-solving and assistance, and are, more than members of many professions, treated by others as a collective (e.g., “tell the nurses to watch his fluid intake”). As a result, our practices are potentially more visible to each other than is the case for professionals whose work is more solitary.

There is a relatively small literature on the topic of professionalism and peer regulation in nursing. Supples (1993) investigated nurses’ responses to substandard practice, and reported that nurses tended to describe these responses in terms of “helping”, i.e., as efforts to understand and arrange help for nurses who are perceived as having difficulty. Staff nurses were often reluctant to confront colleagues, and tended to shift responsibility for this to supervisory personnel. Nursing administrators, meanwhile, were conflicted. As professional leaders, they perceived an obligation to uphold and enforce professional standards, while as mid-level managers, they felt an obligation to protect the interests of the organization (Supples, 1993).

In DeMarco’s (1998) review of several studies on “constructive confrontation” among nurses, she reported that nurses perceived an ethical dilemma in these situations, in which commitments to maintaining relationships conflicted with commitments to truth-telling.

More recent accounts of nurse-to-nurse relations often emphasize the destructive elements, e.g., oppression, intimidation, “bullying” among nurses, and allegations of a generalized culture of psychological brutality and horizontal violence (Baltimore, 2006; Duddle and Boughton, 2007; Hutchinson, Vickers, Jackson, and Wilkes, 2006; Johnson, 2009).

The organization of nursing practice on general hospital units remains an area of relatively scant research. While there exists a basic vocabulary for care delivery models (e.g., primary nursing, team nursing, total patient care), it is not clear to what extent these are being implemented in reliable or consistent ways (Minnick, Mion, Johnson and Catrambone, 2007). Nor is it clear how effective these models are in organizing and producing quality nursing care (Lookinland, Tiedeman and Crosson, 2005).

Research on Magnet hospitals emphasizes the “collegial” relationships in such workplaces, but this element generally refers to collaboration between nurses and physicians, not among nurses themselves (Armstrong and Laschinger, 2006; Kramer, Schmalenberg and Maguire, 2004). While “working with clinically-competent colleagues” and “control over nursing practice” are identified as essential elements of the Magnet hospital model (known sometimes as “Magnetism”), relationships with nursing peers are not explicitly addressed, and “autonomy of nursing practice” is highlighted instead. Meanwhile, over in the world of quality improvement, “crucial conversations” among co-workers and the creation of a “culture of safety”, in which errors and near-misses can be safely acknowledged, have been identified as important tools for reducing errors and enhancing patient safety (Ehrich, 2006; Maxfield, Grenny, McMillan, Patterson, and Switzler, 2005; Moore and Putman, 2008).

A recent synthesis of ethnographic research on hospital nursing identified a set of common elements or domains of nursing practice across multiple studies (Allen, 2007). Taken together, these common elements form a picture of nursing practice that is quite different from that conveyed by the most prominent nurse theories. Allen calls this the “mismatch between nursing’s professional ideals and the workplace experiences of most practitioners” (Allen, 2007, p. 46).

Methods

The results described here are a subset of the data from a larger ethnographic inquiry (Padgett, 2006). The overall purpose of the study was to explore and describe the informal systems of professional regulation among the nursing staff at the unit level. This paper focuses on the specific research question of how staff nurses distinguished between differences in practice that were concerning or troubling – and would therefore prompt a conversation between peers – and those differences that were perceived instead as reasonable or acceptable variations, not concerning, and therefore not prompting a conversation. By “differences in practice”, I mean variations in the delivery of care to patients, how nursing was “done” in very concrete terms. In the interviews, I deliberately left this definition fairly open, so as not to influence participant responses.

The study design was critical ethnography, focused on everyday practices of nursing care, with attention to institutional structures, policy contexts, and surrounding cultural discourses (Crotty, 1998; Schwandt, 1997). Negotiated-order theory and organizational-communication studies provided additional tools for analyzing institutional contexts (Fine, 1984; Miller, 1994; Strauss, Schatzman, Bucher, Ehrlich, and Sabshin, 1975).

The idea of a critical ethnography is intended here to suggest two broad goals. One, the description of cultural practices should be linked to descriptions of political practices and historical relationships of power. Two, the subjective experiences of the participants, or more specifically their reports of those experiences (what they say about them), should be understood in the context of the social conditions as well – neither taken at face value nor rejected out of hand, but viewed as clues to be pondered. Mills’ concept of “vocabularies of motive” (1940) offers a complementary formulation of this approach to the interpretation of participant accounts.

Human-subjects approval was obtained from the University of Washington Human Subjects Division, and from the hospital’s research committee. Informed consent was obtained prior to individual interviews. Unit staff also had the opportunity to “opt out” of the observational part of the study by notifying the researcher; none chose to do so. The researcher had no prior relationship with the unit staff.

The setting for the study was a busy 35-bed general medical unit in a large urban hospital in the western United States. The nursing staff was primarily RNs, with one or two unlicensed nursing assistants per shift. The unit was recommended to me by the hospital research committee in part because it was generally well-regarded, considered to provide high-quality care, and had a fairly stable staff and management. Data were generated through participant-observation, semi-structured interviews, and policy analysis.

Field observations were done on the unit an average of 8 hours a week for a period of six months during 2006, at different times of the day and night, and were oriented toward understanding ordinary ward activities. The purpose of these observations was to provide opportunities to understand the ordinary operations and work routines of the floor, to examine the structural organization of the unit, and to be able to make explicit some of what the participants took for granted (Allen, 2004).

A total of 26 formal, semi-structured interviews were conducted; 19 of these were with staff nurses; four with managers or assistant managers, two with nursing educators, and one with a ward clerk. Interviews were conducted before or after work times in conference rooms or other off-unit locations. Length of the interviews ranged from 30 to 90 minutes. The focus of the interviews was on perceptions of quality in nursing care, relationships with co-workers, and the management of similarities and differences in practice. With permission of the participants, the interviews were tape-recorded and transcribed onto computer files. These transcripts were compared to the audiotapes in order to ensure accuracy, and identifying information was removed. In these excerpts, deletions are indicated by an ellipsis […]; substituted or paraphrased text is indicated with brackets [co-worker’s name].

Results

As noted above, the original goal of this part of the project was to identify how staff nurses distinguished problematic or troubling practice variations from unproblematic (or benign) ones, as signaled by the felt need for a conversation between co-workers. What participants told me, however, was that such conversations were extremely uncommon. They believed that any conversation with a co-worker about differences in practice was likely to be perceived as a criticism, not likely to be taken well or to accomplish much, and therefore considered too risky. The research question therefore changed from “when and how do the staff have these conversations?” to “why do they not have these conversations?” or more specifically, “how do the yaccount for not having these conversations?” (Another question – why did I as a researcher expect the staff to have these kinds of conversations? – is potentially interesting, though beyond the scope of this paper [Padgett, 2006].)

The nurses in this study offered a variety of explanations, often overlapping, for why they did not engage in such explicit conversations. In what follows, I will first describe the organization of nursing care on this unit, drawn from fieldwork observations, interviews, and policy analysis, and summarized in Table 1. The features of nursing practice described here are not believed to be unusual for a hospital unit of this type, but for the purposes of this analysis it is important to make them explicit. After that, I will present two clusters of explanations offered by the staff; these are labeled interpersonal and organizational, and are summarized in Tables 2 and 3.

Table 1.

Work arrangements relevant to the discussion of differences in practice

Helping each other with routine tasks and physical work was expected and necessary – and largely voluntary and informal “Standards of practice” were identified primarily as a management responsibility rather than professional domain – important, but informally balanced with the nurse’s personal preference, experience, and other factors “My patient, my shift, my call”: nurses exercised broad discretion for decisions about nursing care, and cross-shift coordination was mainly voluntary and ad hoc Unit-level “group practice” systems were limited; there was very little in the way of unit-level decision-making, peer evaluation, care plans, or language about shared practice

Table 2.

Interpersonal explanations

I don’t like it, I’m not comfortable with that Other people won’t like it, they’ll be uncomfortable They’ll get mad at me Bad things could happen to me (reprisals) Bad things could happen to the floor (the staff group) I don’t want to be that kind of person – a complainer, a trouble-maker, an unkind or difficult person

Table 3.

Organizational explanations

I have no authority over someone else If it’s my patient, it’s my decision It won’t do any good; people won’t listen We don’t have a group practice There isn’t agreement about standards Management won’t back you up

Conditions of nursing practice on the unit

Here I will outline some of the key features of the conditions of nursing practice on this unit, based primarily on my observations on site. These conditions form a crucial part of the context for understanding the explanations offered by the nursing staff for their actions.

Most of the time, the unit was characterized by a relentless “churning” of patients and staff. Patients were constantly being admitted, discharged, or transferred; there were almost never any empty beds, and a nurse who discharged one patient was very likely to get another. Staff were also in a near-constant state of flux. Most nurses worked 12-hour shifts most of the time, but there were many exceptions to this pattern. Some staff regularly worked 8-hour days or evenings; chronic shortages of staff during the study period meant that charge nurses were always looking for nurses to stay late or come in early; and fluctuations in patient census meant that staff requirements might go up or down for some 4-hour block. Charge nurses spent a great deal of time on the phone and in conversation with staff, trying to recruit people to stay late, come in early, or come in on their days off; schedules were frequently juggled and deals made in order to fill these slots.

The nursing care model used on this unit was what is usually called “total patient care” (Tiedeman and Lookinland, 2004). Nurses were each assigned a group of patients, and were broadly responsible for the nursing care of those patients during their shift. Transfer of information and responsibility across shifts was done through verbal, one-to-one reports between the oncoming and outgoing nurses. Nurse-to-nurse hand-offs occurred as often as 4 times a day; the main change-of-shift reports took place at 7 am and 7 pm, with smaller reports at 3 pm and 11 pm. Coordination of nursing tasks and plans was largely improvised, and relied on informal consensus, suggestions, and persuasion. While physician orders were treated as mandatory (they could be and were challenged, but they could not be ignored), there was no corresponding system for nursing orders. Nursing care plans were written on all patients, but these were treated by most nurses as a regulatory requirement, not as an actual guide to individual patient care. Within the broad guidelines of hospital policy, decisions about care practices were largely considered a matter of the nurse’s individual judgment, not as collective decisions or standards, and even hospital policies were widely ignored on certain issues. Nurses on each shift therefore had broad discretion for decisions about the care of their patients.

This deference to the autonomy of each nurse co-existed with expectations of mutual assistance. Staff who were less busy at any given point were expected to help those who were more busy, and nurses who were more experienced or knowledgeable were expected to help less experienced staff. Nurses who were perceived as not doing their share – not helping enough, or asking for too much help – were scorned and mocked, though usually not directly. This assistance included physical care (lifting, turning, cleaning) and covering each other for meal breaks, as well as advice about navigating the system, sharing information about medications and procedures, and helping to assess or manage deteriorating patients. “Moral support” was very commonly exchanged as well, and indeed this was part of what marked this unit (to most of the staff and to outsiders) as “one of the good ones”: a workplace in which people got along well and supported each other. In the interviews, nurses commonly described their co-workers as their first and most important resource in daily work. While they did not describe this as a perfect system, they consistently reported generally high levels of cooperation and collaboration, and described this as one of the most important aspects of the work environment.

There was almost no formal organization of nursing practice on the unit. There were policies, of course, and periodic staff meetings, several unit-based committees, and a vertical structure, but there was no lateral decision-making system or professional practice model. Significant policy decisions were made either by managers or by hospital-wide committees.

In summary, each nurse was assigned a group of patients, and was given a fair amount of discretion in the care of those patients during his or her shift. At the same time, the effective functioning of the unit as a whole depended heavily on extensive and informal cooperation. An intricate and careful dance was therefore required in order for this system to operate as smoothly as it did; a dance I call mutual deference.

Explanations for the Reluctance to Talk with Colleagues

When asked in the interviews why they did not usually engage in discussions of differences in practice, these nurses offered two sets of reasons. The first set (Table 2) I call interpersonal reasons; the second set (Table 3) I have classified as organizational reasons. (It should be noted that the language in the tables is mine, not based on quotes from participants.)

The first set of reasons were largely focused on psychological or interpersonal effects of such conversations, were they to happen. For example, one nurse said,

I’m not very good at confronting people, to be honest… I don’t like conflict between people. I just don’t. I don’t like to deal with that… I don’t like the discomfort.

In their simplest form, these explanations went like this: open discussion of differences would mean conflict; conflict would mean unhappiness and discomfort; and unhappiness and discomfort would be bad for individuals and bad for the unit. Sometimes this was explained in reference to the idea of being nice – that it wasn’t nice to criticize or complain about people’s work, or to suggest (however diplomatically) that their work was below the standard.

I think a lot of us feel like we’re supposed to be the nice ones… It’s very difficult to say something to some nurses without them taking it personally… In my experience, a lot of nurses will be upset because they think that you’re telling them that they’re not doing a good job… Nurses want to be good, always, and they don’t want criticism.

Unsurprisingly, when they did offer feedback, the nurses said they were very careful how they framed it; they described their efforts in ways that emphasized their attentiveness to these social conventions and relationships. One fairly new nurse, for example, was especially sensitive to the way her feedback might be taken by more senior staff:

I think everyone has their own style of nursing, and I’ve learned that there’s really no right or wrong way… So I don’t disagree, say, ‘oh, no, no, that’s not right.’ But if I know that a certain thing should not be done, then I’ll say, ‘well, you know, I was looking in the [policy manual]’, and kind of, you know, [say it] in a nice way – because no one likes to be told that they’re wrong.… Or I’ll just kind of hint in some way. I mean everyone has that ego, so it’s really hard, telling someone that what they’re doing is not right. Especially when they’ve been doing it for such a long time.

Many of the nurses used very similar language, emphasizing that feedback, if given, would need to be said in the right way – “not harsh”, “not in an accusing way.” However, the simplest and perhaps most common strategy described was to not say anything at all. “I just swallow a lot,” said one nurse. “I just let it go – it’s not worth it,” said another. Even when they were telling me of an incident in which they had spoken up, they emphasized the exceptional nature of it, thereby underscoring the social norm of non-interference.

The second set of reasons offered by the nurses (summarized in Table 3) refers less to interpersonal relationships, and more to working conditions and the structure of nursing practice on the unit. It should be noted that these reasons were not offered by different staff than the first set, but instead are often taken from the same interviews at different points. In other words, participants offered multiple (and sometimes conflicting) explanations of their behavior.

For example, the nurse whose interview is first excerpted above went on to distinguish between conflict with patients and conflict with co-workers, and to say that it was really the second category that was difficult:

I’ll confront patients before I’ll confront nurses… I’ll deal with a patient conflict, because that’s a whole different thing. Because I don’t know them on a personal level. You know, they come in and out. Whereas people here, you know them, you’ve got to give report... So you have to be cordial… Plus, we need each other. Like last night, we didn’t have any [nursing assistants], so you had to ask another nurse to help you to turn patients.

In this excerpt, the nurse moves through both categories: first, the interpersonal relationships (transient in the case of patients – “they come in and out, I don’t know them”, and more durable in the case of co-workers), and then the practical consequences, the effect on work relationships (“you’ve got to give report… you [have] to ask another nurse to help you”).

In an interview with another nurse, we were talking about “following” a nurse (i.e., taking over the care of a patient from a nurse on a preceding shift) when there has been a conflict with that patient’s family, and how, as the next nurse on duty, one has to deal with the fallout from that:

Interviewer: So do you ever have a conversation with the [preceding] nurse about how that went and what they might have done differently?

Nurse: Most of the time, probably not. If I have a relationship with them… If I know the nurse, and they seem like they’re receptive. Because nurses, you know, some you can approach; some you can’t. They’ll become defensive right away if you try to say something. So if I know them, and if it’s a family that I know, and there are certain things that push their buttons and stuff. Yeah, I would try to communicate, especially if the other nurse brought it up, or you’re talking about it, and [you] say, “yeah, this is what you should do, don’t do that.” But other nurses, too, they have their own way of doing things, so of course I’ll be totally dismissed, because they’re like, “this is how I do it” and that’s it.

A bit later, the nurse continued:

I guess the thing is, too, it’s like competing authority. Because sometimes they have – they know what’s right, and so you’re kind of competing about what you’re trying to do for this patient. They have their way of doing it and you have a way you’re trying to do it. There can be a conflict there.

Interviewer: And you don’t have any authority over a co-worker.

Nurse: None at all, none at all. Especially… with the management that we have here, we really feel it’s just not going to go anywhere.

In this excerpt, this nurse also offers a series of explanations, first interpersonal and then increasingly organizational: the other nurse will be defensive; I’ll be dismissed if I challenge their practice style; we’ll be competing over what’s best; I don’t have any authority; management won’t do anything about it. This nurse also offers a set of conditions in which such a conversation could take place, though the conditions seem progressively less likely: if I have a relationship with the other nurse; if I know the family well and know their particular concerns; if the other nurse brings it up, or if we’re already talking about it.

In a final example (taken, as it happens, from my very first interview), a nurse struggled to identify a realm of shared practice in which differences could be considered a common object for discussion, something over which several nurses would have a shared authority:

Interviewer: Say you had one idea about how things should go and somebody else had a different idea. Has that ever happened?

Nurse: I’m trying to think of – I’m sure it has, and I’m trying to think of a specific example. Because if it was my patient, it would be my final decision, so I’m trying to think if – maybe as charge nurse if there was an issue that would come up. I can’t think of a specific issue. […] I guess it depends on how important, what kind of thing it was, too.

The conversation then turned to various examples of what she considered acceptable variations in practice. With some probing for me, she described various conditions in which something might be said, for example if the other person were a student or a new nurse, or if there had recently been a change in the protocol (such that the other nurse might not know). As with the earlier interviews, the possibility of direct conversation sounds increasingly less likely than the alternative, which is to find some way around it:

Nurse: You might see something that was like, ‘well, I wouldn’t do it that way.’… I mean, I know it comes up all the time, but I just don’t know that there’s been a situation where people really disagreed so much that, you know, you couldn’t find a – you know, ‘this is what the [resource] nurse said’, or ‘this is what that person said, so we’ll do it this way’… I just can’t think of anything like that.

The difficulty this nurse had in coming up with examples of this sort was very common; almost every nurse interviewed had many examples ready-at-hand of acceptable types of practice variations, and often struggled to come up with any examples of unacceptable variations. What this nurse refers to here as “a situation where people really disagreed so much” is clearly meant to describe an extreme situation; her preferred way of resolving such a conflict (implied in the fragmented statement above) is to invoke a third party with more authority, a nurse designated by the hospital as a resource person, for example.

Discussion

Fieldwork observations illuminated how much staff nurses depended on each other for practical assistance, and for support in large and small ways. There was no formal system for organizing or allocating that assistance and support, however. Maintaining a positive spirit of cooperation and mutual aid was highly valued; at the same time, the “autonomy” of each nurse was also highly valued. These two principles – cooperation and autonomy – were reconciled in practice by giving each other wide discretion in decisions about nursing care practices, what I have called “mutual deference.”

There are two ways of describing this concept. The more positive way would be to say: I will respect the judgments of my fellow nurses, and I expect them to give me that respect in return. The more negative framing of this principle might be: You leave me alone, and I’ll leave you alone. The practice of mutual deference allows for offering help, but primarily when it is asked for, or is necessary to protect patients – though this last category is unclear, and suggests a very high standard, sufficient to overcome the ordinary expectation that each nurse is allowed to make her own choices about practice. By contrast, the realm of shared responsibility and decision-making (i.e., horizontal accountability or group practice) seemed vanishingly small.

To initiate a conversation about nursing practice, therefore, is to raise questions about the specific practices of another nurse, and would be seen as an act of criticism and challenge, potentially threatening to the group order, divisive and dangerous. In the absence of either a structure for, or a clear cultural concept of, professional collegiality, there was little room for the discussion of practice differences to be perceived as anything other than interference at best, and at worst, a personal attack – not nice or helpful, nor likely to improve care.

Nurses’ reluctance to engage in collegial negotiations and conflict around standards and performance has been reported before; this study offers a troubling confirmation of this pattern. A collective unwillingness to engage in the activities of professional self-regulation on the ground level – via peer monitoring, dialogue, and as needed, conflict and confrontation – leaves the profession without one of its most important tools for maintaining accountability, and for protecting the public from substandard practice.

What is different here is the explanation that I am proposing for this reluctance. Previous studies have tended to emphasize the social psychology of nurses (e.g., conflict avoidance, fear, and intimidation) or defects in nurses’ communication skills. This study calls attention as well to the organization of work as a relevant factor. Relations among nursing co-workers – which professional ideology would identify as collegial relations – were organized instead in a loose and semi-voluntary “teamwork” model, in which the framework for cooperation was primarily interpersonal rather than professional. Without a language for professional group practice (denoting shared decision-making and responsibility), the staff on the unit identified themselves simply as people who helped each other and got along well. The need to collaborate in the delivery of everyday nursing care, without any organizational recognition of nurses’ lateral relationships, resulted in those relationships being constructed on interpersonal terms.

Meanwhile, the “professional” element of nursing practice was associated more with the (alleged) autonomy of the individual clinician. On the everyday level, this was translated as your shift, your patient, your call. This concept of professionalism, however, seems to work against collegiality and peer accountability.

The idea of a “vocabulary of motives” (Mills, 1940) can be useful here. Mills offers a view of motives, or “reasons for doing things”, that focuses on their status as linguistic transactions rather than merely internal feelings. Mills notices that such reasons are offered (usually in response to a question, such as “why…?”), and therefore function as social events and contextualized explanations. A “motive” is something that is recognizable by other people in that situation (this includes the recognition of one’s own motives in a particular context). Reading these nurses’ accounts through this framework allows us to see the problem as sociological, and not merely psychological. That is, what nurses lack in this situation is not so much an intrinsic desire to collaborate with their colleagues as a socially recognized vocabulary for communicating that desire. In this site at least, there were few opportunities for nurses to discuss “shared practices” (how we do things here) that could be seen by others as an activity that was ordinary, safe, and reciprocal, rather than judgmental or intrusive. This analysis suggests that nurses may not need “tools for constructive conflict” (as is often suggested) as much as they need stronger (i.e., more readily recognizable) reasons for having those kinds of conversations – about the “common ground” of practice, in which we can talk about how we do things here.

Since nursing professional care models generally emphasize the relationship between an individual nurse and an individual patient (Allen, 2007), the ways in which nursing practice is delivered by groups of nurses to groups of patients have been largely invisible to researchers, managers, and nursing theorists (though see Latimer [2000] for a noteworthy and important exception). Staff nurses engaged in this work cannot afford to overlook this fact, however, and are highly attuned to (perhaps even hyper-vigilant about) their relationships with co-workers.

In the context of the ubiquity of current reports of nurse-to-nurse “bullying” and lateral violence, it was interesting that these nurses did not bring this up as a concern. They did describe co-workers as being more or less responsive to feedback, more or less pleasant to deal with; certain staff were more likely to be considered mean or rude, but there were not the reports of pervasive and persistent patterns of fear and intimidation that have been described elsewhere.

Limitations

It is important to note that there is no direct evidence from this study that more collegial conversations, of the type proposed here, would necessarily improve practice. Because the primary focus on the study was on a single unit, it was not possible to explore the role of unit management, for reasons of both confidentiality and lack of comparison cases. More generally, the results of this study should be read with the ordinary cautions of a single report from the field. Every effort was made to select a reasonably typical unit, and to pay special attention to elements that might be more generally relevant across clinical settings. Nonetheless, the ways in which this unit was or was not representative of the broader world of nursing practice cannot be ascertained from within the study itself, and readers should evaluate its relevance to their own settings to decide whether these findings ‘ring true’ for them (Sandelowski and Barroso, 2002).

Implications

The research described here suggests that professional collegiality and horizontal accountability are inhibited by the combination of multiple factors:

• the needs of nurses to get along with each other and exchange practical assistance;

• the emphasis on “autonomy” over “collaboration” in professional discourse;

• the uncertain scope of responsibilities and authority for individual nurses, and

• the uncertainty about the shared responsibilities: what can one nurse expect or require of another?

• the lack of a lateral system for organizing practice and structuring accountability among the nurses on the unit: to what extent does care need to be consistent across shifts and providers, and how is that to be accomplished?

The results of this study suggest that we take more seriously the issues of teamwork and mutual deference, as well as the disincentives for actions perceived as personal challenges and threats to harmony. If we wish to shift unit-level practice to include more open and direct communication with colleagues, it is crucial that we attend to the pragmatic considerations that influence nurse-nurse relationships. In this context, it might be worthwhile to re-visit the issues of care delivery models and systems (Latimer, 2000; Lookinland et al., 2005; Minnick et al., 2007).

It might be useful as well to investigate more fully the ways that nurses accomplish the work of collaborating in the delivery of care, and how “getting along” and “not stirring up trouble” are part of that. Instead of seeing staff nurses as merely handicapped by poor communication skills and anxieties about conflict, we might consider them also as highly skilled and subtle communicators, who navigate complex interactions and systems, largely successfully.

This study also confirms the value of critical ethnography as a research methodology, especially for investigating the everyday practices and working conditions of nurses and nursing. Critical ethnography enables us to address structural relationships and organizational factors, and to connect what people say with what they do, and the conditions in which they do it (Tope, Chamberlain, Crowley, and Hodson, 2005).

Finally, these findings call attention to the tension between nursing’s preferred images of itself and the more complicated situations on the ground. This tension takes two forms. In the first, the preferred image of nursing as an independent profession comes up against the situation of nurses as employees of large institutions, in this case hospitals. In the second case, the image of nurses as individual clinicians providing care to individual patients comes up against the more common situation of staff nurses in collective practice: the delivery of care by groups of nurses to groups of patients. In both cases, we have tended to prefer the simplified images (professional autonomy and individualized care) rather than the more complicated situations, in which we are entangled in institutions and with each other.

As an ethnographic report, this interpretation presumes “that nurses’ actions should be understood as a rational response to the constraints of practice” (Allen, 207, p. 45). However, ‘rational’ does not necessarily mean desirable. Shirley (2007) argues that nursing tends to conflate professional autonomy, patient autonomy, and patient advocacy, combining them into one big ethical imperative of “individual choice.” This strategy serves us poorly in all three domains, not only by confusing them with each other, but also by limiting our ability to explore interdependency and the multiplicity of moral claims, and to “build a more complex moral environment” (Shirley, 2007, p.23).

Staff nurses readily invoke the languages of professional autonomy and individualized care in describing their practice. The most readily available discourse for describing their collegial relationships, however, remains interpersonal: being nice, being helpful, and getting along. While these are admirable and useful traits, they seem not quite enough for a full professional vocabulary.

What is already known about the topic?

• Collegial relationships in nursing are considered crucial for ensuring quality and safety, and for maintaining the standards of professionalism.

• Like other professionals, staff nurses tend to avoid conflict with peers, and especially to avoid open discussions of errors and problems.

• Explanations for this avoidance have tended to concentrate on psychological and cultural factors, with relatively less attention paid to organizational influences.

What this paper adds.

• There are significant practical incentives for maintaining positive relations with co-workers, and corresponding disincentives for questioning or challenging the practices of others.

• The mutual dependency of staff nurses leads to a system of mutual deference, a strategy of reciprocal tolerance and non-interference that gives wide discretion to each nurse’s decisions about care practices.

• Efforts to change staff behaviors that focus only on communication skills and becoming “comfortable with conflict”, without addressing the practical organization of work, are therefore unlikely to be effective.