End-of-life Care and Interprofessional Communication : Not Simply a Matter of " More "

INTRODUCTION A well-functioning interprofessional team has been identified as a central requirement for high quality palliative care. In particular, interprofessional communication and teamwork have been directly linked to patient and family health outcomes. However, evidence suggests that substandard communication and team collaboration between healthcare providers is a persistent challenge that is heightened during palliative care in in-patient settings. This research examined the mechanisms of communication that shaped and impeded interprofessional team practice and coordinated palliative care on acute medical and long-term care units. METHODS This participatory action research project was informed by planned-action and educative-research strategies. The research team worked with healthcare practitioners who cared for dying people in acute and long-term care settings to develop and change practices and institutional arrangements through concurrent phases of ongoing analysis, dialogue, action, and reflection. Data-gathering methods included audio-recorded baseline interviews, observations and on site interactions with field notes, focused group discussions, and meetings. All data was coded using NVivo 9 and a subsequent second level analysis was conducted using Critical Discourse Analysis and Relational Inquiry as an analytical framework. FINDINGS Two main findings included (a) the way in which participants drew upon socio-cultural knowledge to structure and enact communication processes and to describe and interpret their communication experiences within the team, and (b) four recurring relational disjunctures in which conflicting and/or competing messages, goals, or processes hindered the flow and processes of communication and interprofessional team practice. CONCLUSION Given the way in which ideologies and normative practices shape and contribute to ineffective communicative patterns, the findings suggest that the issue is not only how much communication is happening, but the nature of that communication. Thus, addressing the conflicting and/or competing messages, goals, or processes shaping the flow and processes of communication within the interprofessional team is necessary. In particular, explicitly addressing the complex interplay between autonomous professional practice and interprofessional team collaboration is a crucial step in supporting more effective communication and team cohesion. Received: 04/24/2012 Accepted: 06/13/2012 Published: 09/24/2012 © 2012 Doane et al. This open access article is distributed under a Creative Commons Attribution License, which allows unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. H IP & End-of-Life Care ORIGINAL RESEARCH 1(3):eP1028 | 2 Introduction Research has shown that high-quality palliative care requires a coordinated interprofessional team approach. However, while interprofessional communication and teamwork have been directly linked to patient and family health outcomes (Leathard, 1994), evidence suggests that substandard communication and team collaboration between healthcare providers is a persistent challenge (Bokhour, 2006; Bronstein, 2003; DeLoach, 2003; Reese & Sontag, 2001; Wittenberg-Lyles, Parker Oliver, Demiris, & Courtney, 2009). This challenge is particularly heightened during palliative care in inpatient settings. Building upon previous research findings that illuminated the need for more coordinated palliative care in acute medical and long-term care (LTC) settings (Bern-Klug, 2004; Stadjuhar & Davies, 2005), a participatory action research (PAR) project was undertaken to actively engage interprofessional team members in addressing this need. Included in this research was an investigation of the communication patterns that were occurring within the interprofessional team, with a particular emphasis on examining the discursive practices, social relations, and established patterns of interaction and the impact of those processes on palliative care goals.


Introduction
Research has shown that high-quality palliative care requires a coordinated interprofessional team approach.However, while interprofessional communication and teamwork have been directly linked to patient and family health outcomes (Leathard, 1994), evidence suggests that substandard communication and team collaboration between healthcare providers is a persistent challenge (Bokhour, 2006;Bronstein, 2003;DeLoach, 2003;Reese & Sontag, 2001;Wittenberg-Lyles, Parker Oliver, Demiris, & Courtney, 2009).This challenge is particularly heightened during palliative care in inpatient settings.Building upon previous research findings that illuminated the need for more coordinated palliative care in acute medical and long-term care (LTC) settings (Bern-Klug, 2004;Stadjuhar & Davies, 2005), a participatory action research (PAR) project was undertaken to actively engage interprofessional team members in addressing this need.Included in this research was an investigation of the communication patterns that were occurring within the interprofessional team, with a particular emphasis on examining the discursive practices, social relations, and established patterns of interaction and the impact of those processes on palliative care goals.

Literature Review
Palliative and hospice approaches have highlighted the way in which high-quality palliative care requires expertise from different sectors and disciplines.Fundamentally an interprofessional practice, a palliative approach is dependent on coordinated care between interprofessional team members (Connor, Egan, Kwilosz, Larson, & Reese, 2002;Ellingson, 2003;Reese & Sontag, 2001;Wittenberg-Lyles, Parker Oliver, Demiris, & Regehr, 2007;Wittenberg-Lyles, Parker Oliver, Demiris, & Regehr, 2009).Specifically, interprofessional collaboration has been shown to improve patient outcomes, such as symptom control and self-determination at end-oflife (Hearn & Higginson, 1998), as well as nurses' and physicians' job satisfaction (Manojlovich, 2005).Other identified benefits of interprofessional teamwork include increased resource-efficiency and innovation, holistic care, and the opportunity for care providers

Implications for Interprofessional Practice
• Improving communication in interprofessional healthcare teams is not simply a matter of creating mechanisms for 'more' communicating but also examining the ideologies and normative practices that shape how communication is organized and structured and how they contribute to ineffective patterns of collaboration.
• An examination of the contradictory processes of autonomous professional practice and team collaboration needs to be explicitly undertaken to determine the impact of this potential disjuncture within specific contexts of healthcare practice.
• Everyday talk within professional subgroups and the larger interprofessional team can reinforce role distinctions and foster a pattern of inclusion/exclusion.Thus, the way individuals and subgroups distinguish and identify themselves and each other needs to be considered to ensure such distinctions are not circumventing effective interprofessional practice.
• Better communication and interprofessional team collaboration in palliative care requires an examination of how individuals are identified as knowers/unknowers and how these distinctions shape information sharing and knowledge translation.
• Differences within interprofessional groups and the challenges they give rise to (such as the relational disjunctures of professional autonomy/team collaboration, inclusion/exclusion, knowers/unknowers and time/priorities) offer potential sites to enhance team collaboration and communication.to develop new skills and approaches (Firth-Cozens, 1998;2001).
While research shows that the majority of people spend their final days in inpatient settings (Heyland, Lavery, Tranmer, Shortt, & Taylor, 2000), acute medical wards and LTC units are particularly problematic places to die (Formiga, Olmedo, Lopez-Soto, Navarro, Culla, & Pujol, 2007;Kayser-Jones, 2002;Kayser-Jones et al., 2003;Stajduhar & Davies, 2005;Willard & Luker, 2006).Part of the challenge lies in how practice is organized and structured within those settings.For example, practice in acute medical units is paced and organized around acute illness trajectories with a focus on curing, while practice in LTC has historically been organized around convalescence and rehabilitation models.Thus, palliative care is not an explicit aspect of the practice models that dominate those settings.Moreover, the way in which such practice is organized and enacted is not well aligned with coordinated palliative care planning and/ or interprofessional collaboration to support the dying process (Formiga et al., 2007;Kayser-Jones, 2002;Kayser-Jones et al., 2003;Stajduhar & Davies, 2005;Willard & Luker, 2006).
Since effective communication has been identified as the glue that both holds the interprofessional team together and facilitates coordination of care (Baggs, Norton, Schmitt, & Seller, 2004;Hearn and Higginson, 1998;Jones, 1997, Meier & Beresford, 2008;Mills, Neilt, & Dunn, 2008;Reader, Flin, Mearns, & Cutherbertson, 2007), further examination of communication within interprofessional teams needs to be undertaken.While there is a wealth of literature that speaks to the challenges of inter-professional communication, less well understood is the impact of existing patterns of communication on palliative care.Specifically, improved communication during palliative care in acute and LTC settings requires further study.

Method
This PAR project was informed by planned-action theories (Graham et al., 2006) and educative-research strategies.Planned action involves deliberately engineering changes and ways of doing things in social systems (Graham et al., 2006), and educative research refers to a collaborative inquiry process that focuses on the examination and transformation of ideas and practices through dialogue and action (Hartrick, 1998;Smith, 1993).Combined, these two approaches supported the intent of the project-namely, for the core members of the research team to work with healthcare practitioners who cared for dying people in acute and LTC settings with the goal of developing and changing ideas, practices, and institutional arrangements through concurrent phases of ongoing analysis, dialogue, action, and reflection.This core research team had professional practice backgrounds in nursing, social work and psychology.
The research took place over a two-year period.Specific data-gathering methods included a series of audiorecorded baseline interviews, a series of observations with field notes, rapid response interviews (consisting of 3-5 brief questions to get specific kinds of data based on our ongoing analysis), focused group discussions, and meetings.The observations and group discussion and meetings were conducted and recorded in field notes.The first four authors participated in the datagathering.All data was coded using NVivo 9, a computer software package used for organizing and grouping data into 'sets' which can be compared, linked and contrasted.The codes were then reviewed by the four researchers to ensure inter-rater reliability.Approval for the research was obtained from the University and Health Region's Human Research Ethics Board.

Baseline Data Gathering
The PAR project began by conducting 69 baseline interviews with participants across interprofessional fields including clinical nurse educators (CNE), registered nurses (RNs), resident care assistants (RCAs), chaplains, allied health practitioners, and physicians in both acute medicine and LTC settings.A total of 68 individual interviews and one five-person focus group were completed at baseline.The interviews focused on having participants describe their current experiences in providing palliative care within these settings, including their concerns and goals for improvement.All participants worked within the same urban health region in Western Canada.The intent of the baseline interviews was to identify and clarify the specific concerns and needs related to improving palliative care in their respective settings.All baseline interviews were audio recorded, transcribed verbatim and coded using NVivo 9. Comparisons of transcripts coded by team members identified areas of similarities and differences.Transcripts were read over several times to identify recurring, converging and opposing themes, key concepts and illustrative examples from the data.The researchers then reviewed the coding to confirm interrater reliability.

Action Cycle Data Gathering
As baseline data analysis was being undertaken, we identified two specific sites within the health region for participation in a more concentrated data collection and intervention process (referred to in PAR as action cycles).The first site was an acute medical unit in a large urban hospital; the second was a LTC facility with a mixed population of geriatric residents, complex neuro patients and patients in an activation program.The PAR cycles were focused on targeting and addressing specific concerns related to palliative care on the units.
The first step of the action cycles was to conduct a series of observations, rapid response interviews and focused group discussions on each of the units.During these activities researchers gathered information about the particular units including the priority concerns and goals of the team members.Field observations included buddy shifts, attendance at workshops, and medical rounds, staff meetings, and physician meetings.These observations included an interactive process in which the researchers asked questions, elicited further information about observed interactive patterns, and clarified understandings.As data were gathered through these mechanisms it was recorded in field notes and subsequently reviewed and coded.Through this continuous process of data gathering/data analysis, preliminary concepts and themes were further developed and categorized.This analytic process enabled data to be continually reviewed by members of the research team and shared with the practitioners, and coding categories to be continuously revised and refined.
As we conducted this ongoing analysis, communication consistently emerged as a pivotal element that was having a significant and far-reaching effect on the process of palliative care.Participants described a "lack of communication" as central to the gap between real and ideal deaths on the units (the gap between what was occurring and what they believed should be occurring).
There was an ongoing assertion that "if only people could learn to communicate with each other, problems could be identified, discussed, and resolved." People often spoke of needing to "get on the same page" to achieve more coherence as a team and a more informed and coordinated approach to palliative care.
What we as a research team found intriguing about this claim was the way in which the 'problem' of communication was narrated as a deficit.That is, there was an assumption that the interprofessional team members were "not communicating" and that what was needed was "more communication." Yet although experientially people were feeling they needed "more" communication, we observed and our data confirmed that there was in fact a lot of communication happening.The problem did not seem to be about a lack of communication but more about the nature and patterns of communication that were in play.That is, communication was occurring and messages were being conveyed between the members, but the nature and content of those messages and the existing interactive processes were not necessarily supporting the experience of team connection, nor were they supporting the outcome of coordinated palliative care that team members were wanting.Given this finding, we undertook a more targeted analysis of communication in the hope that a clearer knowledge about existing communication patterns and processes could inform interventions during the action cycles.

Analytic Framework and Data Analysis Process
Our initial analysis had highlighted that the "problem of communication" was not a simple matter that could be addressed by better mechanisms of information sharing or other technical solutions (for example, getting people talking more or providing education about "communication skills").Rather, a much more complex interworking of people, roles, differing knowledge-bases, power relations, and so forth seemed to be at work.The communication challenge was analogous to what Heifetz (1998) has termed an adaptive challenge.Heifetz distinguishes technical challenges (in which it is possible to clearly identify a problem, a clear solution, and the knowledge or skill set needed to address the challenge) from adaptive challenges (ones that lie within people and situations and require something beyond the incorporation of knowledge, technical skills, or solutions)."Technical problems reside in the head; solving them requires an appeal to the mind, to logic, and to the intellect.Adaptive challenges lie in the stomach and the heart.To solve them, we must change people's values, beliefs, habits, ways of working, or way of life" (Heifetz & Linsky, 2002, p. 35).While most problems pose a combination of technical and adaptive challeng- es, Heifetz (1998) asserts that the biggest error made in efforts to effect change in action (for example, change in communication processes) is to identify the technical aspect and apply technical means without attending to the adaptive elements.Heeding Heifetz's warning, we decided that the values, ideologies, structures, and normative practices that were shaping communication practices required further examination if intervention was to be successful.
To tease out and better understand the relational interworkings that were shaping the experience and process of communication, we drew upon critical discourse analysis (CDA) and relational inquiry (Hartrick Doane & Varcoe, 2005).Fairclough (1993) defines CDA as: [D]iscourse analysis which aims to systematically explore often opaque relationships of causality and determination between (a) discursive practices, events and texts, and (b) wider social and cultural structures, relations and processes; to investigate how such practices, events and texts arise out of and are ideologically shaped by relations of power.(p.

135)
We observed that messages were not only sent through language but were often embedded in material and technological artefacts, images, written texts, social groupings, and so forth.CDA includes "analysis of the dialectical relationships between discourse (including language but also other forms of semiosis, e.g.body language or visual images) and other elements of social practices" (Chiapello and Fairclough, 2002:185).However, to complement our discursive examination, we employed relational inquiry.Relational inquiry focuses attention on the relational assemblage-on how seemingly disparate elements intersect and converge (for example, assumptions, values, people, normative patterns and practices, social/power relations, systemic factors, past experiences, and so forth (Hartrick Doane, & Varcoe, 2005;2007;2008)).
Informed by CDA and Relational inquiry, we oriented our analytic process toward action by purposefully juxtaposing the descriptions team members offered with the communication actions we observed.Employing a discursive/relational lens, we analyzed the different texts to systematically "lift out" patterns and arguments (LeGreco & Tracy, 2009, p.1532), identifying the embedded beliefs, assumptions, ideologies, normative practices, and relational patterns.Ricoeur (1991) describes how in moments of action it is possible to see the interplay of ideology-how people are scripted into roles, social relations, and communicative patterns.Within this examination we considered three levels of interacting 'text'-the micro (e.g.everyday talk), the meso (e.g.organizational structures), and the macro (e.g.cultural norms such as social hierarchy).Our orienting questions included: How is communication organized and enacted?What is shaping and informing existing patterns of communication?What are the consequences of existing communication practices?

Results
As we intentionally stepped out of an either/or binary understanding of communication-the perception that the practitioners were either communicating or not communicating-and turned our attention to the way in which communication was happening, two major findings were highlighted.The first major finding was the way in which ideologies and socio-cultural givens shape and give rise to everyday communication processes.The second major finding was what we came to understand as 'relational disjunctures.' These relational disjunctures involved juxtapositions of conflicting and/ or competing messages, goals, or processes that were shaping the flow and processes of communication.We describe the elements of these two findings in such a way as to elucidate the relational interplay between them.
Relational Disjuncture One: Autonomy and Team professional purview and attended to his/her individual patient assignment.Yet, although care was organized and structured in this autonomous way, the practitioners continually talked about the importance of "team" and "teamwork." Everyday conversations touted 'teamwork' as being crucial to good care and as the remedy for addressing the confusion and lack of consistency often experienced during palliative situations.In essence, teamwork was an ideological 'truism' that showed up in any conversation related to high quality care and good communication.The quote below offers an example of how, when concerns over good decision-making or palliation arose, the conversation quickly turned to a discussion of how teamwork was both the problem and the panacea.
" [U]nderlying all that we're doing is that team-building process.Because the stronger, more cohesive a team, the better the communication, better knowledge is shared, the more consistent, the more buy-in.I mean you know, all of that." This juxtaposition of teamwork and autonomy was highly significant in structuring and enacting communication.While the everyday discourse positioned teamwork as an either/or-either people worked as a team (and communicated) or they did not work as a team (and did not communicate), in observing communication/team work in action it was possible to see a far more complex relational process at play.For example, while the nurses worked autonomously it was evident that they simultaneously worked very much in concert with the team's normative practices, coordinating their actions with the larger structures and processes that organized how the team functioned.To illustrate, the nurse in the above quote coordinated her work with other team members by quite literally getting out of the way when they came in the room.She created space for the other team members to "do their own thing." Although not necessarily consciously, the nurse was both practicing autonomously and participating as a team member by operating in accordance with established norms.Paradoxically, the way to work as a team was to work autonomously.That is, team members related with and to each other by staying out of each other's way so each could do their respective jobs-ensuring they did not cross each other's boundaries unless absolutely necessary.
These 'team' messages that were sent and decoded be-tween team members, even in the absence of verbal articulation, reflected a highly significant form of communication.Within the current structure of autonomous practice, teamwork was only really sanctioned when it was not possible to proceed autonomously.Thus, there was a disjuncture between how teamwork was being enacted and the type of teamwork the everyday discourse purported as necessary for good palliative care.Given this disjuncture, it is not surprising that experientially people felt they needed "more" teamwork.The way practice was organized and the patterns of interaction between team members, including the way they moved and related to each other within and through interpersonal spaces of care, served to reinforce the gap between them.
Relational Disjuncture Two: Inclusion/Exclusion (In or Out of the "Sisterhood") And I think the nurses often share information, a kind of sisterhood, the RNs.It doesn't always filter down to the LPNs and the Care Aides.And LPNs have access to the chart so when we're sharing it and you don't write it down, then they get frustrated that then they have to ask or can't find it themselves.
The second relational disjuncture we found was that of inclusion/exclusion.When the topic of 'teamwork' was evoked within the everyday discourse, it was often not clear who the 'team' included.For example, on the medical unit there was a lack of clarity around whether 'team' referred to everyone (RNs, LPNs, physicians, allied health workers) or consisted of a group of specialized individuals within their roles (for example the nursing team).When participants were directly asked to clarify what/who constituted the team, they replied that as a result of role divisions and the way those divisions shaped interactions, different forms of teamwork were enacted in different situations.For example, on the medical unit there was the strong "home team" of nurses and the larger "less cohesive" team that included nurses, physicians, and allied health workers.On the LTC site there was sometimes the team of RNs, physicians, and allied health workers, where RCAs were excluded, and at other times discussions of team referred to RNs, LPNs, and RCAs, since physicians were not present on a daily basis.At the same time on the LTC site there was what was referred to as the "sisterhood" of RNs.This inclusion/exclusion disjuncture significantly affected how individual practitioners saw themselves, the confidence and knowledge they had, and how they enacted their roles.While the idea of 'sisterhood' may be experienced as positive for those within it, in terms of the larger team it clearly delineated insiders from outsiders.This relational inclusion/exclusion strongly impacted communication processes in terms of who spoke to whom, what the different dyads or groups talked about, how they related around residents, their access to information, and so forth.In addition, the inclusion/exclusion disjuncture served to solidify role divisions and structure communication processes according to roles.For example, on the medical unit RNs were often excluded from rounds with the physicians, who preferred to speak directly to the charge nurse.
Even though the RNs had more direct knowledge of the particular patients and the interchange of their knowledge with the physician's would have been helpful, that interchange often did not happen.In response to this pattern, one RN stated what was required was "nurses stepping up to the plate in terms of pinning the physicians down and we have to be more aggressive to say, 'Listen I want to know what's going on here and are you thinking this is end of life and can you order end of life orders' and the physician not brushing us off." Similarly, RCAs described being locked out of the team work between the RNs and the physicians, describing a palpable sense of being segregated at the "bottom of the hierarchy." The paradox within this inclusion/exclusion dynamic was that it simultaneously hindered teamwork and strengthened it.For example, on the medical unit it thwarted fuller relations within the larger team on one hand, and on the other it solidified relations in the smaller team of nurses.Being a knower or unknower also shaped whether a team member was informed, whether and how much time was allotted to them to communicate, the nature and amount of communication that occurred, and which resources they had access to.Everyday examples included distinctions about who gave report to whom and in what form, who charted where, who was invited to attend clinical care meetings, who had access to educational opportunities, who waited to speak to whom, and so forth.For example, as mentioned earlier, on the LTC unit RCAs were often excluded from clinical care meetings, from charting, and from educational opportunities.

Relational
Interestingly, this knower/unknower distinction was experienced and perpetuated all the way down the hierarchy.While in some cases RNs expressed distress that their knowledge was not being valued by physicians or administrators, RCAs often described the way in which RNs wielded the same kind of knowledge authority over them.The relational disjuncture was the way in which people were simultaneously designated as knowers and unknowers.For example, while RCAs were expected to have the knowledge to manage and respond to resident needs (knowers), because they were seen as "only an RCA" (an unknower) they often were not given important information about a resident nor were they formally acknowledged as knowers in the same way as others on the healthcare team (for example, they were not able to chart what they knew about the resident in the formal health record).This relational dynamic also played out between RNs and physicians.For example, physicians often did not inform RNs about advance care plans for patients so that the nurses were at times left in the dark as to the specific direction care was taking.Yet at the same time, the physicians experienced frustration when the nurses did not meet their expectations in terms of being a knowledgeable practitioner when they phoned for information about a patient.As one physician described, "When they phone, I want information relevant to the care I am giving, not some rambling report." Thus, to effect better communication and team collaboration in palliative care, the assumptions and distinctions about knowers/unknowers and the processes for information sharing and knowledge translation need further consideration.The fourth relational disjuncture we found on both units was that of time and priorities.Within the fastpaced world of contemporary healthcare, where discourses of "limited resources" and "doing more with less" dominate, the juxtaposition of time and priorities was palpable.As described in the quote above, as the practitioners were "stretched" across competing priorities, they often felt unable to do what they needed to do to meet their professional obligations in the way they deemed sufficient.Paradoxically, the organizational cutbacks meant they were doing more, yet at the heart of their practice they found themselves doing less for the patients they cared for (in terms of both quality and substance).

Relational
This disjuncture between time and priorities was acutely evident in the communication processes.The everyday discourse and experiential reality of practicing within highly restrictive time pressures served to structure and organize communication in particular ways.For example, the combination of patient acuity, workload, and staffing ratios left little time for conversing about patients, and team members found it increasingly challenging to coordinate time to discuss patient issues and/or develop a coordinated plan of care.This served to both marginalize and align individuals and subgroups.As members were positioned according to role distinctions, they were subsequently allotted varying levels of "legitimacy" to engage in professional talk.
For example, a physician could request an answer from anyone on the hierarchy (access their time), while an RCA (according to the RCAs) did not have the same privilege.Many RCAs described how in seeking information about a resident or wanting to discuss care with an RN they were often "not even heard." At the same time, as illustrated in the quote below, even when a team member wanted to be supportive the information flow was such that they were isolated by competing demands and the frenetic pace.
I know we had one episode where I didn't realize that the LPN was really struggling for the whole day that she was caring for a palliative person and she didn't realize that she could draw the RN in to help and that's what part of the RN's role is, is to see when the LPN is struggling and help...And there was no team going on there so nobody realized that this LPN was struggling with a very difficult palliative patient and she needed help.
This relational disjuncture of time and priorities involved a double bind of wanting more collaboration yet achieving less.Moreover, this relational pattern and pace of practice served to fuel communication tensions and power relations on each of the units.This was particularly evident in situations that the team members referred to as "bad deaths"-times when the patient suffered.Team members continually told "bad death" stories of distressing palliative situations where, because of relational dynamics, they had been unable to affect the outcomes for their patients.Interestingly, the experiential impact of these distressing situations was felt by everyone, regardless of where they were on the hierarchy.For example, although physicians had the power to both structure communication and decide the content and the timing of that communication, they described the way in which their best laid plans could be circumvented by a nurse being out of synch with the plan.Again, limitations of time and competing priorities appeared to heighten these occurrences.

Discussion
Previous research has identified both the importance and challenge of inter-professional collaboration and communication in palliative care (Baggs, et al., 2004;Hearn & Higginson, 1998;Jones, 1997;Meier & Beresford, 2008;Mills, Neilt, & Dunn, 2008;Reader, Flin, Mearns, & Cutherbertson, 2007).Examining interprofessional team communication from a different vantage point than previous research, this study moves beyond the either/or binary-the perception that practitioners are either communicating or not communicating.Specifically, the study offers insight into the relationships among ideologies, relational disjunctures, and communication practices.In so doing, it reveals the tensions that may hamper effective communication in interprofessional teams.
Ricoeur (1991) has described the way in which "the interpretive code of ideology is something in which men (sic) live and think...it operates behind our backs, rather than appearing as a theme before our eyes.We think from it rather than about it" (p.251).Similar to Anselm et al's (2005) finding that the Canadian healthcare system, with its busy work schedules, staffing ratios, and frantic pace of practice, is itself a barrier to communication in palliative care, this study reveals the relationships among system level ideologies, normative practices, and communication patterns.As Papadatou (2009) has described, working in a chronically underresourced sector, [C]are providers engage in situations that lead to ongoing over-activity and over-agitation as they move frantically from one stressful event or critical episode to another.Such over-agitation is not a temporary response to increased job demands, but a permanent condition that results from the teams' attempts to avoid anxiety-provoking relationships, threatening circumstances, suffering, and death.
Care providers attend to individuals and families, to team problems or conflicts, to bureaucratic or administrative issues, and even to secondary tasks that absorb all their energy with an acute sense of urgency.They overinvest in work tasks and leave no space or time to invest in relationships with people.(p. 231) This fast-paced system of care played a central role in shaping the relational disjunctures that were enacted and experienced by the practitioners who participated in this study.Overall, the findings of this study reveal the way in which the inter-professional team members drew upon their socio-cultural knowledge to structure and enact communication processes and to describe and interpret their communication experiences within the team.Of significance is how this background ideology/knowledge-influenced role distinctions and served to shape interpretations and behavioural responses between team members-including the ways in which they normatively conformed to existing practices and also went about challenging those they deemed problematic (Hall, 1980).
At the heart of the relational disjunctures, and the communication pattern they gave rise to, was the precedence of the biomedical hierarchy.Previous research has described the ways in which physicians tend to control interpersonal communication and how this control is partially explained by the elevation/dominance of biomedical knowledge in the medical care context (Wittenberg-Lyles et al., 2007;2009).Similarly, in our research, team members' communication was shaped according to the way they were deemed knowers/unknowers within the biomedical knowledge hierarchy.
Although the negative impact of status differentials and hierarchical splits within healthcare teams has been documented (Cott, 1997;Kahn, 2005), our research reveals how this hierarchy is experienced and the subsequent impact of that experience.Similar to Thompson's (2003) findings, team members often felt undervalued, "demeaned, " and/or "invisible." Often the RCAs (in the LTC setting) or the RNs (in the medical setting) possessed important information about patients because of their ongoing interactions with them.Yet because they were positioned as unknowers, that information did not receive the same attention or value as that of team members positioned as knowers.Moreover, at times they were not given information that could have facilitated the care process.For example, on the medical In a social-network analysis of team structures in a specialized, multi-level care facility, Cott (1997) found that: [W]hile teamwork may be increasing the participation in decision-making by health professionals other than medicine, rather than flattening the hierarchical structure throughout the health care division of labour, its effects are limited to a group of higher status professionals.A clearly defined hierarchy remains for the lower status subdisciplines and "I decide, you carry it out" has simply become "We decide, you carry it out." (p.1418) Cott contends that this difference in status is attributed to differences in structure and work, purporting that physicians and allied health practitioners are mainly involved in decision-making and problem-solving and nurses are mainly involved in carrying out the tasks once the decision has been made.While certainly many (ourselves included) would argue that nursing involves far more than task completion, the decision-making/ problem-solving aspect Cott points to and the underlying assumptions about who does what are highly significant when thinking of palliative care where illness trajectories are uncertain.For example, a patient with cystic fibrosis waiting for a lung transplant has a fluctuating illness trajectory that can change quickly.In situations like this, dying is ambiguous (the patient may die or may continue to live for several more years); thus, health practitioners often struggle to know when to initiate and/or 'do' palliative care.Since nurses are most often the primary care givers, their knowledge becomes vital-they are the ones who are there to identify fluctuations and respond to them, to make the judgement that a different intervention is required, and to engage the inter-professional team members' expertise for this complex care planning.Yet, as Thompson (2003) has reported and our study demonstrated, nurses often find themselves having difficulty getting physicians to respond to their clinical assessment that, for example, more symptom control is needed.
Previous research has described the way in which everyday talk between health practitioners not only transfers information, but also enables individuals to construct particular professional relations, negotiate responsibility, and establish common values (Atkinson, 1994;Hunt, Benford and Snow, 1994).For example, Li and Arber (2006) found that the telling of 'atrocity stories' by nurses in team meetings in three palliative care settings functioned to emphasise the boundaries between professional groups and enable the groups to voice their complaints.Story-telling within a group was used to uphold group members' own moral integrity and also to resolve tensions in the team context.Similarly, in a discourse analysis and ethnography of a medical ward in Britain, Allen (2001) found that nurses engaged in specific types of story-telling, and that the rhetorical form of their 'atrocity' stories functioned to delineate a moral division between nurses and other medical staff.More recently, in their study of professional identity formation, Lingard, Reznick, DeVito and Espin ( 2002) described how the use of 'othering' narratives (consciously or not) served to simplify and shape the interpretations of 'other' team members and their actions.As in this previous research, it was evident in our research that everyday talk served to both contest and solidify the knowledge hierarchy and inclusion/exclusion communicative patterns.In addition our study revealed how the relational disjuncture between everyday talk and communication practices affected the way individual practitioners saw themselves, the confidence and knowledge they had, and how they enacted their roles.The everyday talk reinforced role distinctions and fostered the development of shared frames of interpretation such as 'team communication /no team communication.' This overall pattern of relational disjunctures illuminates the intricacies involved in improving communication and coordinated interprofessional teamwork.Hall and Weavers (2001) contend that "two issues are emerging in healthcare as clinicians face the complexities of current patient care: the need for specialized health professionals, and the need for these professionals to collaborate" (p.867).The relational disjunctures reveal the tension between those two essential elements.While holism is upheld as a virtue in patient care (Sulmasy, 2002) Elias (1978;1982) has described the way in which people are configured within a mesh of social relations through which they learn to discipline and control their own bodies in ways that are socially prescribed and mirror the social habitus of their group.These unconscious habits of conduct form into dispositions (ways of thinking, patterns of acting and relating) and become ingrained in bodily responses and behavioural actions.

Conclusion
Combining CDA and relational inquiry, our study highlights the ways in which dominant ideologies are lived out in embodied, normative practices that perpetuate and reinforce particular communication patterns.These findings challenge the dualism of communicating/not communicating that shapes most discussions of communication in healthcare settings, illuminating the ways in which communication in some form is always happening.Consequently, improving communication and team functioning is not simply a matter of creating mechanisms for 'more' communicating.While technical interventions might be helpful, enhancing communication and team collaboration in palliative care also requires consideration of the ways in which current practices-including verbal and non-verbal signs, silences, omissions, inclusions, and exclusions-perpetuate particular relational patterns.Highlighting the complexities that underlie and shape communication processes, the findings suggest that enhancing communication within interprofessional teams requires careful examination of the ideologies and normative practices impacting the team members' perceptions and experiences of themselves and their colleagues, as well as their understandings of teamwork.
A particularly significant aspect of the findings is the interface of autonomous practice and team collaboration.
While the everyday conversations of teams seemed to have embedded within them the assumption of a team as an integrated machine (as a Fordist model, a group with a common vision, implicitly without conflict), given specialized professionalization it would seem that this understanding is insufficient.Teams need to be constituted with a clear understanding of the strong historical, emotive, philosophical, and socio-cultural roots that underpin the divisions between the professions (Puntillo and McAdams, 2006) and the tendency for cross-disciplinary misunderstanding and defensiveness between disciplines to arise (Connor, et al., 2002;Rafferty, 1996;Reese and Sontag, 2001;Wicks, 1998).
While teams typically come together because they share a common goal or purpose, a team can become even "tighter" if there is a perceived common adversary who interferes with their ability to achieve their goal.
Given this complex interplay between autonomous professional practice and interprofessional team collaboration the first step to support more effective communication and team cohesion would be to explicitly recognize and address the tension.As Watts (1975) has observed, while boundaries may be seen as lines that separate, they are also sites of joining.Offering the example of the shoreline, where land and water touch, he asserts that although the line between them may distinguish and demarcate where one ends and the other begins, that boundary also marks where and how they join and unite (Watts, 1975;Wilbur, 2001).Rather than ignoring the differences within interprofessional groups and the adaptive challenges and relational disjunctures they give rise to, the disjunctures need to be purposefully enlisted-called into question and worked with.
In particular the interplay of autonomous practice and interprofessional collaboration needs to be examined within specific contexts to enable more explicit identification of the tensions and disjunctures that might be interrupting cohesive teamwork.One step would be to begin paying attention to the everyday talk within professional subgroups and the larger interprofessional team.How is the discourse inadvertently fostering a pattern of inclusion/exclusion through engrained role distinctions?Another might be to notice the way individuals and subgroups are distinguishing and identifying themselves as knowers or unknowers and how these distinctions are shaping information sharing and knowledge translation.Finally, taking a more concerted look at how to strategically enlist the complementary knowledge and capacities of the different professionals to address the very real time constraints and competing priorities could serve to enhance both team collaboration and cohesive end-of-life care.Questions interprofessional teams could consider include: How might the processes of autonomous practice and team collaboration work in concert and support each other?How might the boundaries between the professions become the actual sites of interprofessional collaboration?Such an examination would enable consideration of how autonomous practice and cohesive interprofessional teamwork might be more effectively interwoven to enhance end-of-life care.
often found themselves caring for dying patients and interacting with the family members without knowing the physician's specific advanced care plan and/or what the family had been told by the physician.