Navigating the Hierarchy: Communicating Power Relationships in Collaborative Health Care Groups

Negotiating power relationships

This study uses the lens of text and conversation (Taylor et al., 1996) to examine how health care groups negotiate the tension between the established hierarchy and collaborative models of care through everyday communication and interaction. Conversation is the actual interactive process through which organizing occurs; through conversation, organizational members construct and negotiate (or co-orient around) texts, which form a kind of “conceptual scaffolding made up of words, phrases, turns of speech, metaphors, anecdotes, all of which are there because of the distillation, stored in language in the memory of participants, of their personal and collective history of previous interactions” (Taylor, 1999, p. 26). Texts are created in contextualized moments of conversation, but once created, they can exist outside of the conversations through which they were created. Putnam and Cooren (2004) further explained this process and its significance: “Since texts can endure through memory traces, documents, and signs, they form a way for interactions at the local level to be reproduced apart from their original production” (p. 325). As texts are reproduced over and over again through patterned conversation, they gain a kind of hybrid agency that they share with human conversational participants, which means that they play a role in shaping future interaction. This paper argues that power relationships in health care organizations are a product of how groups negotiate intersecting texts through their conversations (Taylor, 2001).

The texts that sustain hierarchical power relationships exist in the form of written documents that codify power relationships (e.g., legal mandates that require most other providers to be supervised by a physician) as well as memory traces that group members carry with them from one interaction to the next. Laws are relatively fixed texts, but like any other text, they were created through historical conversations among physicians and others who sought to preserve physicians’ authority (Freidson, 1974). The texts that shape power relationships have persisted across time and space through imbrication—a process through which texts produced in previous interactions intersect with new texts to become the taken-for-granted content that actually shapes future interactions (Taylor, 2001). Imbalanced power relationships often hinder successful team collaboration, and they have consequences for patient care. But there seems to be some variation in how different health care groups navigate the texts that shape these power relationships through conversation—either in ways that exacerbate the tension between the hierarchy and collaborative models of care or in ways that reduce this tension. Two questions guide this study:

Observation

Observation sessions focused on patient rounds as a critical part of the collaborative care process. A total of 20 observation sessions were conducted across the six care groups; each session typically lasted between one and two hours depending on the length of patient rounds for each care group. ² To set up an observation session, the researcher’s point of contact would connect her via email with someone who worked with the group to arrange the days and times for observation. The number of observation sessions conducted with each group depended on the researcher’s ability to develop an understanding of the standard rounding process ³ that took place within that group.

Detailed field notes on each observation session focused on the time and location of rounds, who was present, where people sat or stood, who led the discussion, who participated and how they participated, how the discussion unfolded, what the process was like, and what kinds of interactions took place among group members. If the group conducted bedside or walking rounds, notes were initially written in a small notebook and then transcribed into digital form immediately after the observation session. If the group conducted office rounds, notes were taken on a laptop. These observation sessions resulted in 70 single-spaced pages of field notes.

Interviews

The researcher also conducted 49 interviews with care group members—including nurses, physicians, and ancillary staff from each of the six groups. ⁴ Interview questions focused on: participants’ professional background and their role in providing care, ⁵ group and organizational culture, ⁶ collaboration during ⁷ and outside of rounds, ⁸ and other barriers or factors that might impact patient care. ⁹ Interviews had to be relatively short to accommodate the busy schedules of the interviewees. On average, they lasted between 15 and 20 minutes—though they ranged from five minutes to an hour. The semistructured interview format allowed for flexibility, and the researcher highlighted priority questions in advance to prepare for interviewees with very little time. All interviews except for one were conducted in person at the hospital in locations identified as convenient by the interviewees (e.g., the café, on the units in empty consult rooms, in private offices, etc.). One interview was conducted over Skype because the interviewee was on a leave of absence from work. Interviewees were assured of confidentiality and asked if recording was okay for research purposes. Only two interviewees asked not to be recorded, so extensive and careful notes were taken during those interviews.

(Senior) Doctors make the decisions

Physicians openly claimed their power to make decisions. Many physicians from across all of the care groups voiced this power in the interview data. For example, in describing her role in the PICU care group, an attending physician explained that when she was on service, she was the “lead physician in terms of decision making for everything that comes in and out of the unit—for everything that’s done on the unit.” When asked what she would do if another provider made a patient care decision she disagreed with, she said, “Well, in my position I guess I am the one that gets to finally make the decision.”

This text was also evident in stories told by interviewees about group interactions and in the observation of rounds when (senior) physicians sanctioned, ignored, or denied the decisions of others. In one example during BMT rounds, the team discussed a patient who had vomited in the night. The social worker joined the conversation, noting that the patient had told her nurse that there may be things she should talk about with a psychologist. The social worker then said that her intern would see the patient the next day. At this point the attending physician jumped in to indicate that he was “okay with the social worker’s intern seeing her.” Despite the fact that psychosocial care was the social worker’s primary domain and was not the physician’s area of expertise, the doctors-make-decisions text gave him the authority to sanction the social worker’s decision. The act of sanctioning reinforced the power imbalance between medical and psychosocial providers. An exchange during Hem/Onc rounds demonstrated how physicians sometimes ignored decisions made by other disciplines. The group was discussing one of their patients who was receiving physical therapy without a physical therapist present during rounds:

Hem/Onc attending: He needs a break from PT. He’s not really into it right now.

Hem/Onc resident: PT came to his room because they needed a new order.

Hem/Onc attending: Yeah, it’s fine that they’re coming, but . . .

Hem/Onc resident: . . . he just needs a little break, yeah.

The doctors-make-decisions text allowed the physicians to temporarily stop the patient’s physical therapy without consulting a physical therapist. In another example, a PICU attending described how she sometimes had to simply deny the decisions of junior physicians. “There has been a rare instance where I really disagree with the senior fellow, and I have to say, ‘I’m sorry; it just has to be this way’ after like hours and hours of discussion.”

Nurses and ancillary staff also used this text in deferring to physicians. A Hem/Onc staff nurse shared that rounds were the most important time of the day for communicating about patient care plans because “that’s where the attending is there, and they are the ones who pretty much- they are the big guys. They make the decisions.” A GenPedSurg nurse practitioner (NP) used the text in her explanation of how involved she felt in the discussion during rounds:

I feel equal in terms of making suggestions or saying “Hey, have we considered this aspect?” or “Hey, what do you think about this kind of situation?” And I feel like it is taken under consideration in a respectful way, but then whatever the attending or the fellow decides—that’s what we go with.

A nutritionist from the CVA care group similarly described her deference to physicians’ decisions in a hypothetical scenario where she might disagree with the medical team. She concluded by saying, “So I guess I would provide my professional opinion and leave it to the doctors or the NPs to decide what to do.” In this example, the nutritionist deferred not just to the physicians in her group but to the nurse practitioners as well, recognizing their legal and professional ability to diagnose patients and prescribe medication that led some to perceive their role in the group as similar to that of physicians. This perspective highlights the complexity of group power relationships since, unsurprisingly, most physicians did not share this perspective. In another example during BMT rounds, the team debated why a patient was not eating well. The staff nurse caring for the patient that day was present. She gave her opinion to the attending physician on how to resolve the food issue based on the 12-hour shift she had spent at the patient’s bedside but then deferred to the physician, saying, “but it’s up to you—what you think is best.”

Unlike the doctors-make-decisions text that was openly articulated in interviews by representatives of all disciplines and care groups across the organization, the other texts that reinforced hierarchical power relationships were rarely articulated openly—perhaps because they clashed with the organization’s codified values. They were primarily identified in patterns of interactions from stories told in the interviews and from the observation data, and they seemed to shape conversations, and power relationships, differently across the six care groups.

(Senior) Doctors have the most knowledge and expertise

Another text that reinforced hierarchical power relationships through group conversation can be explained by the assumption that (senior) doctors have the most knowledge and expertise of anyone in the care group. Two of the most common ways this text shaped power relationships in conversation were through teaching and research. A few ancillary staff members mentioned teaching as part of their role, but the researcher never actually observed teaching that involved any members of the team except physicians. In Hem/Onc rounds, if the team had down time while waiting for the attending physician to arrive, the fellow would typically offer to lecture on a topic chosen by the residents. During these lectures, the non-physician members of the group would chat quietly amongst themselves or look at their phones. These teaching moments certainly shaped power relationships between senior and junior physicians, but they also shaped power relationships between physicians and other members of the care group who were typically not invited to participate in the educational experience. This text also manifested in conversations about research. While people from across different disciplines conducted various kinds of health services research, large medical studies were typically authored by physicians even though other members of the team helped with data collection protocols. In one example, a BMT attending hospitalist used research to draw a distinction between his knowledge and expertise and that of a nurse practitioner on the team who had a very similar role as the attending in the patient care process. During rounds the attending discussed protocols for a research study with the rest of the group. He said, “If there’s a patient who’s consented to [a particular treatment] and has [a particular condition], just send an email to me and [the research administrator] to let us know, and we’ll tell you what to do.” The nurse practitioner (NP) asked, “Do I get to be an author on the paper?” and the hospitalist responded in a way that reinforced the NP’s exclusion from authorship: “Of course—if you want to help write it.”

Physicians also used this text to challenge the authority of other members of the care group. In one example from the interview data, a CVA charge nurse described a new rounding process called care progression rounds, which was focused on discharge planning. A consulting company hired by SCCH had decided that charge nurses should run these care progression rounds, and she explained that the physicians in her care group were not happy about this decision. One physician told her, “I don’t understand why you’re leading it when all you say is room number, the patient’s name, and that’s it . . . when I can just say that and give you the whole background and diagnosis.” Junior physicians often negotiated their power relative to nursing leaders through knowledge and expertise.

(Senior) Doctors are the most vital members of the care group

Care groups also reinforced hierarchical power relationships through the use of another text rooted in the assumption that (senior) doctors are the most vital members of the care group. One way this text shaped power relationships was through physician preferences, which were often privileged over established group processes. In one example, it was the first day that a new group of residents started their rotation in the PICU. A fellow gave them an introduction explaining that the fellows rotated each week and emphasizing multiple times that “every fellow does things differently and has different preferences for how rounds will go, so they need to expect that.” In some groups, the individual preferences of physicians affected group processes from day-to-day so that others had little control over their schedule and their time. The CVA interpreter who sometimes filled in on the med surg unit said she felt that:

It’s not as organized [on the med surg unit] where I know for the day I am going to be doing all these things. They [med surg doctors] call us when things happen . . . It’s harder when I cover [that unit] . . . since I’m so used to a more organized pattern.

She described the challenge of being at the beck and call of physicians with little control over her own schedule and the other work she had to get done. A GenPedSurg fellow noted that only half of the attendings from his/her ¹⁰ care group chose to attend rounds, which could create more work for him/her, especially when the team received “inappropriate consults.” S/he said that when the attending comes to rounds:

then all we have to do is put in our note that it’s seen and examined by the attending . . . and then it’s done, but if I say it, then they are calling and asking about- “well should we do such and such?” and that’s not appropriate. I don’t need to waste my time on that.

In addition to creating confusion and more work for other members of the team, a lack of clear team process could disempower others and leave them uncertain of their role in the group, which could make them less likely to speak up or challenge physicians or to do so only in a polite or careful way. One BMT charge nurse described a culture in her care group not of challenging physicians but simply of asking questions because that is “when you learn and when you understand the rationale.” A BMT staff nurse similarly noted, “I will ask why that way instead of another way. I will try obviously not to say it in a mean way.”

Physician preferences also often determined spatial norms in ways that shaped group power relationships. Two of the care groups (BMT and Hem/Onc) conducted table rounds in an office instead of walking rounds on the unit, and one group (GenPedSurg) conducted a hybrid that started with table rounds and progressed to walking rounds. For office rounds, the groups typically sat around a long table and discussed each patient’s status instead of visiting the patients at the bedside. By choosing to do office rounds, physicians made it harder for nurses to attend rounds, which often took place while they were in the middle of their morning tasks on the unit. Because it took so much time for staff nurses to travel back and forth between the unit and the office, physicians from the Hem/Onc group simply decided staff nurses would no longer attend rounds. Instead the charge nurse from each unit attended rounds. While this may have been more efficient for the physician team, it created more work for others since the charge nurses had to find each staff nurse after rounds and explain the plan of care for every patient or a resident had to call the staff nurses to update them. One resident succinctly described the problem this created:

We would round in the room so the nurse was never there, so we have to call the nurse to give them the plan. But sometimes things get so busy; you don’t call them later, and then they don’t really know what’s going on.

Efficiency for physicians was often used as a justification for physician preferences—even when it made more work for other group members and ultimately made the group as a whole less efficient.

The choice of space for office rounds was also significant as was the arrangement of group members within the space. Hem/Onc rounds took place in the physician work room—a space outside of the nursing unit that “belonged” to physicians and that others needed a door code and permission to enter. Larger groups in particular, like the Hem/Onc group, organized themselves around the long table in ways that reinforced imbalanced power relationships with physicians comfortably seated in chairs at one end of the table while nurses and ancillary staff scrambled to find a seat at the other end of the table or sat on counters or even stood behind the table when those seats were taken. If no attending was present, the fellow sat at the head of the table, though when an attending was present, he or she occupied this powerful position and the fellow moved one seat over. Once during BMT rounds, a doctor from another team poked his head into the office to ask a question about a patient, and the charge nurse actually gave up her chair to this doctor and stood for the duration of his discussion with the team. These spatial norms reinforced power differences among the disciplines and affected participation. Field notes from the first day of observing Hem/Onc rounds noted, “The charge nurse just sits there so far—doesn’t say a word.” Physicians typically led the discussion during office rounds while others listened and occasionally interjected to provide information, but non-physicians had no defined, active role in the process; they were there to hear the physicians’ conversation and to contribute information—especially when directly consulted by a member of the physician team.

We work together for the kids

Participants from all care groups seemed to agree that working collaboratively was essential for accomplishing the shared goal of providing excellent care to patients and families at SCCH. This text was also codified in SCCH’s mission and values. An occupational therapist who worked across all the groups said, “I think overall everyone really likes to share information, and I feel like people are there for the kids.” A PICU staff nurse described the SCCH culture in terms of “the level of respect between all disciplines, and then, of course, our mission that we are here for our children, and I think that really does- is shown in day-to-day actions; it’s not just words.” A BMT attending described her team as rising above disciplinary differences to provide excellent care. “It’s not, ‘I’m a doctor; I’m a nurse.’ It’s like- there’s a team . . . Everybody is responsible.” This sentiment was echoed by a BMT nurse practitioner who said, “I think we work together pretty close, and we would point out mistakes to each other without- At the end of the day, it’s for the patients.” Representatives from different disciplines across all six of the care groups articulated the work-together-for-kids text in the interviews. However, only some of the groups seemed to translate this organizational text into group texts that shaped daily conversations and interactions in ways that challenged the established hierarchy.

All group members have valuable perspectives

One text that groups used to challenge the hierarchy was based on the idea that all members of the group had a unique and valuable perspective to contribute to the patient care process. This text shifted conversations away from a focus on knowledge and expertise, which physicians have so much of from their many years of training, and instead emphasized the unique contributions of different members of the care group that came from their specific training or their unique role in caring for patients. The text challenged power relationships when the group acknowledged and valued the importance of all perspectives by creating an open, informal environment where everyone was able to communicate freely with each other. One way that groups created this environment was by calling each other by their first names. A BMT staff nurse said, “I think we’re a very friendly group, and it’s very easy for the nurses to talk to doctors and the other way around. We call the doctors by their first name here. We don’t say doctor so and so.” A PICU fellow similarly noted that people in his care group really respected each other, and “the ways that I have seen this manifested are people calling each other by their first name across the levels—trying not to have too much of a hierarchy situation; they are willing just to help each other out.”

Another way that care groups created informality and openness was through spatial norms. Unlike some of the other groups, in the PICU, physicians shared ownership of the physician workroom with other disciplines. This shared ownership became clear on the first day of PICU observation:

I met the PICU attending in the hallway by the nursing station and introduced myself . . . Then he walked me down to the “physicians’ workroom.” He kept putting it in scare quotes, and it became clear why when we got there, and he noted that it’s the “physicians’ workroom,” but half of the people in there when we arrived were nurse practitioners!

The PICU and CVA care groups also chose to conduct walking rounds instead of table rounds. Because walking rounds were conducted on the nursing unit, it was much easier for nurses to join rounds. Also, during walking rounds, the group would stand in a circle outside each patient’s room while they discussed the patient’s status and the care plan for the day. The circle could expand or contract to include everyone unlike the table during office rounds, and there was no “head” of the circle.

Non-physicians also used this text to challenge power relationships by asserting the value of their perspective. In one example, a physical therapist who worked across all six teams talked about being “pretty vocal in explaining my point of view and my rationale for why a patient can or can’t do something that they [medical team] would like done at that time.” She demonstrated this in her stories about interactions with physicians as well: “Often the physicians are having their own rounds of patients that are rounds that we aren’t all involved in where they are going room to room. I have included myself in those rounds when I found that it’s important.” A PICU nutritionist actually held formal teaching sessions with the residents rotating through the PICU care group—an arrangement that she advocated for and that physicians in the group had sanctioned, which signaled to junior physicians the value of her perspective. Nurses also spoke up in a variety of different situations to ensure their perspective was heard. In one example from CVA rounds, “a very confident nurse . . . asks for a change in [a patient’s] meds that will apparently make her job easier and seems like it won’t have any effect on the patient’s care.”

Established group processes are important

Another text that groups used to challenge the hierarchy was based on the assumption that established group processes were important to the care process. In groups with clear, established processes for participation and sharing information during rounds, the individual preferences of different physician were less likely to significantly alter the process from day-to-day and the rest of the group knew what to expect. Also, an established process meant that a clear role could be defined for each member of the group. This text was evident in some groups from the organized way that rounds were conducted and from the way interviewees from these care groups described their experience of rounds. In most care groups, the physician trainees (residents or fellows) would “present the patient” (i.e., describe vitals, status changes, medications, etc.) during rounds, but the PICU and CVA groups had decided that the staff nurse would present the patient instead. In his interview, a PICU fellow explained how rare it was for nurses to have this role:

They have the nurses present [the patient report] on rounds, which I have never seen anywhere else, which makes a- it’s a big change or a paradigm shift almost, but it makes the nursing team, I think, much more invested in the care of the patient.

A PICU resident described this group’s rounding process as “more structured” than other groups, and nurses seemed to benefit from this. In an example during PICU rounds, one of the new residents began to present the patient and then suddenly remembered this was the nurse’s job:

The resident assigned to this patient gives his one-liner and then immediately starts reviewing the patient stats, vitals, meds from his info sheet, but then looks up and says, “Oh” . . . [smiling], and the staff nurse steps in and continues with her data . . . on vitals and meds.

In this example, the established process literally interrupted the hierarchy mid-sentence. A staff nurse from CVA also described how “things are more organized” on her unit. CVA staff nurses would prepare for rounds by writing “a brief background, what happened overnight, any events that you want to discuss, and any cleanup orders that need to be done by the doctor” (CVA staff nurse). This preparation allowed them to actively participate in the rounding process by contributing important information from a nursing perspective. The CVA rounding process also empowered nurses throughout the day because, as a charge nurse explained, the physicians would often discuss contingency plans with the staff nurses in anticipation of possible changes in the patient’s status:

So there is always an A plan and a B plan a majority of the time so that if you as a nurse are dealing with one thing and you are trying to work out that A plan, you already know what the B plan is before we even get to that point because you can anticipate it.

This established process signaled physicians’ trust and confidence in the nurses to evaluate changes in the patients’ status and to make minor changes to the patient care plan without always having to contact the supervising physician.