WEEK 7 DISC

ORIGINAL ARTICLE

Roles, processes, and outcomes of interprofessional shared decision-making in a neonatal intensive care unit: A qualitative study Sandra I. Dunna, Betty Cragga, Ian D. Grahama, Jennifer Medvesb, and Isabelle Gabouryc

aSchool of Nursing, University of Ottawa, Ottawa, Ontario, Canada; bSchool of Nursing, Queen’s University, Kingston, Ontario, Canada; cDepartment of Family Medicine and Emergency Medicine, Université de Sherbrooke, Québec, Canada

ABSTRACT Shared decision-making provides an opportunity for the knowledge and skills of care providers to synergistically influence patient care. Little is known about interprofessional shared decision-making processes in critical care settings. The aim of this study was to explore interprofessional team members’ perspectives about the nature of interprofessional shared decision-making in a neonatal intensive care unit (NICU) and to determine if there are any differences in perspectives across professional groups. An exploratory qualitative approach was used consisting of semi-structured interviews with 22 members of an interprofessional team working in a tertiary care NICU in Canada. Participants identified four key roles involved in interprofessional shared decision-making: leader, clinical experts, parents, and synthesizer. Participants perceived that interprofessional shared decision-making happens through collaboration, sharing, and weighing the options, the evidence and the credibility of opinions put forward. The process of interprofessional shared decision-making leads to a well-informed decision and participants feeling valued. Findings from this study identified key concepts of interprofessional shared decision-making, increased awareness of differing professional perspectives about this process of shared decision-making, and clarified understanding of the different roles involved in the decision-making process in an NICU.

ARTICLE HISTORY Received 5 July 2016 Revised 14 November 2017 Accepted 11 January 2018

KEYWORDS Deliberation; interprofessional; neonatal intensive carequalitative descriptive; shared decision- making

Introduction

When patient health issues are complex, a broad range of knowledge and expertise is required to identify the best options for care and improved outcomes. In this scenario, collaboration among professionals is essential (Reeves, Pelone, Harrison, Goldman, & Zwarenstein, 2017; Van Den Bulcke et al., 2016) for both effective care decisions, and in preparation to support patient and family involvement in decision-making for preference-sensitive decisions. The pro- cess of shared decision-making (SDM), as a key component of interprofessional practice (Canadian Interprofessional Health Collaborative, 2010; D’Amour, Ferrada-Videla, Rodriguez, & Beaulieu, 2005), enables the use of separate and shared knowl- edge and skills of caregivers to optimize patient care (Joseph- Williams et al., 2017; Légaré et al., 2011a).

Interprofessional collaboration (IPC) and shared decision- making in the intensive care unit (ICU) has been associated with a decreased rate of patients’ death and readmission to the ICU, nurse and resident job satisfaction (Baggs et al., 1999), improved patient knowledge/understanding, satisfaction and trust (Shay & Lafata, 2017), improved end-of-life care (Puntillo & McAdam, 2006) and reduced adverse event rates (Jain, Miler, & Belt, 2006). Power differentials between professional groups, lack of joint clinical decision-making or poor decision-making processes have been shown to contribute to critical incidents (Grinspun, 2007; Reader, Flin, & Lauche, 2006; Reeves et al., 2015).

Most of the literature on SDM focuses on the dyadic relationship between physician and patient. In this dyadic model of SDM, a clinician (most often a physician) and a patient both participate in the process of decision-making and reach healthcare choices together (Charles, Gafni, & Whelan, 1999; Makoul & Clayman, 2006). Although advocated as an optimal model of treatment decision-making (Elwyn, Edwards, Kinnersley, & Grol, 2000), and considered the pin- nacle of patient-centre care (Barry & Edgman-Levitan, 2012), SDM can be a challenge to achieve (Gravel, Légaré, & Graham, 2006; Légaré, Ratte, Gravel, & Graham, 2008). For example, the conceptualization of SDM, when limited to the physician-patient dyad, does not adequately reflect the current realities of clinical practice where other participants are often involved (e.g. situations where patients are supported by family members, where incompetent or seriously ill patients require proxy decision makers to act on their behalf, or in cases where several physicians are involved in the decision- making process with a single patient). The dyadic SDM model also completely negates the essential roles of other members of the interprofessional team in patient care planning and decision-making and the influence of the environment (e.g. primary care versus intensive care setting) on the decision- making process (Fox & Reeves, 2015; Gachoud, Albert, Kuper, Stroud, & Reeves, 2012; Kon, Davidson, Morrison, Danis, & White, 2016).

CONTACT Sandra I. Dunn sdunn@bornontario.ca Centre for Practice Changing Research, 401 Smyth Road, Ottawa, Ontario K1H 8L1, Canada Supplemental data for this article can be accessed here.

JOURNAL OF INTERPROFESSIONAL CARE 2018, VOL. 32, NO. 3, 284–294 https://doi.org/10.1080/13561820.2018.1428186

© 2018 Taylor & Francis

To facilitate the SDM process when more than one health- care professional (HCP) is involved in care, Légaré and col- leagues (2011a) developed an Interprofessional Approach to SDM Model (IP-SDM) to help guide practice. In the IP-SDM model, different professionals collaborate together to identify best options, and support the patient or family to be involved in decision-making about those options for preference-sensi- tive decisions. The patient and family (or surrogate decision maker), the initiator of the SDM process (any HCP who identifies the health problem and the decision to be made), the decision coach (who is trained to support the patient’s involvement in decision-making), and HCPs involved in care are key roles within the IP-SDM model. As part of the IP- SDM process, stakeholders exchange information about the relative benefits and harms of the options and elicit the patient’s/family’s values and preferences (Légaré et al., 2011a).

Based on the IP-SDM model, members of the interprofes- sional team must reach a state of equipoise – meaning they agree that a decision point exists with one or more potential options, including the option to remain status quo, and for which the benefits and harms need to be weighed across each option (Elwyn, Frosch, & Rollnick, 2009). However, the fact that clinicians form professional opinions based on the research literature, clinical experience, intuition, and ideology, as to the effects of particular treatments, means they are rarely in a state of equipoise, to begin with (McCleary, 2002). How interprofessional teams reach equipoise and when to engage the patient or family in the SDM process is not clearly articu- lated in the IP-SDM model.

Little is known about SDM from the perspective of HCPs other than physicians (Légaré et al., 2008; Visser, Deliens, & Houttekier, 2014) and how to effectively operationalize inter- professional shared decision-making (IPSDM) in settings where patient acuity is high, rapid decision-making is neces- sary, and the patient condition, interprofessional team con- figuration, and the environment are unpredictable, such as in intensive care. In this emergency, high stakes clinical envir- onment barriers to SDM include: time limitations for deci- sion-making (Azoulay, Chaize, & Kentish-Barnes, 2014; Joseph-Williams, Elwyn, & Edwards, 2014; Visser et al., 2014), the diversity of HCPs involved, workflow and lack of continuity of care (Joseph-Williams et al., 2014), poor com- munication between HCPs (Vivian, Marais, McLaughlin, Falkenstein, & Argent, 2009) and relationships that are com- plicated by stressful situations (Vivian et al., 2009). In addi- tion, medical terminology used by clinicians and the ability of patients and families to understand and use the information they receive (Joseph-Williams et al., 2014), physicians’ unease dealing with surrogate decision makers (Visser et al., 2014), HCPs lack of knowledge, training and experience with SDM (Boland, McIsaac, & Lawson, 2016; Légaré et al., 2008), and HCP attitudes towards SDM (Joseph-Williams et al., 2014; Légaré et al., 2008) present challenges for SDM. Interprofessional tensions, and power differentials within the team (Visser et al., 2014; Vivian et al., 2009), and between HCPs and the patient or family (Joseph-Williams et al., 2014) may impede collaboration and interfere with full exploration of care options during SDM (San Martin-Rodriguez, Beaulieu, D’Amour, & Ferrada-Videla, 2005). Finally, lack of triggers to

initiate SDM and engage family members in the process (Kryworuchko, Stacey, Peterson, Heyland, & Graham, 2012), and uncertainty about which clinical situations are suitable for SDM (Boland et al., 2016) are obstacles to SDM in ICU.

Increased understanding of the processes involved in IPSDM, how the members of an interprofessional team in intensive care collaborate to identify the decision to be made and the options for care (the first step in the IP-SDM model), has the potential to improve the quality of the decisions, the support provided to patients and their families (Légaré et al., 2011b), and IPC in general (Haggerty et al., 2003).

The purpose of this study was to explore the nature of IPSDM in a neonatal intensive care unit (NICU) to under- stand the processes involved in decision-making from the perspective of HCPs working on the interprofessional team in the unit, and to determine if there are any differences in perspectives across professional groups. This study is one part of a larger primary study that used an explanatory mixed methods design. The primary study included 4 phases: a systematic review of the literature, a survey of members of the interprofessional team in an NICU (Dunn, Cragg, Graham, Medves, & Gaboury, 2013), interviews with selected members of the interprofessional team and observations of IPSDM during patient care rounds. This paper reports the results of the qualitative interviews and describes the simila- rities and differences in professional perspectives about the key roles, processes, and outcomes of IPSDM in NICU.

Methods

Design and context

An exploratory qualitative approach (Sandelowski, 2010) was used for this phase of the study consisting of semi-structured interviews with members of the interprofessional team work- ing in NICU. Qualitative descriptive studies provide a com- prehensive overview of specific events experienced by individuals or groups and are based on naturalistic inquiry, and a commitment to studying something in its natural state (Lambert & Lambert, 2012).

A tertiary care NICU in Canada was the study setting. This unit provided complex care to approximately 300 infants a year. A NICU was selected as the setting for the study because of unique factors that can both hinder and facilitate IPSDM. These factors include: high patient acuity and instability, the need for coordination of care and collaboration among many different professional groups, a model of practice where the healthcare team comes to the patient rather than the patient coming to see individual health care providers (HCPs) and the need for surrogate decision-making.

Recruitment strategy and participants

Participants included registered nurses (RN), physicians (MD), respiratory therapists (RT), and other health profes- sionals (OHP) (e.g. pharmacists, occupational and physical therapists, dieticians, and social workers). A purposive sam- pling strategy was used to recruit HCPs from all professional groups to participate in interviews to ensure maximum

JOURNAL OF INTERPROFESSIONAL CARE 285

capture of different professional perspectives and facilitate exploration of the common and unique manifestations of IPSDM experienced by the interprofessional team (Palinkas et al., 2016).

The lead author, as previous member of this interprofessional team, was known to the participants. As an insider to this NICU, she had a previously established professional working relation- ships with this team and the benefit of knowing the people, the system of care and the environment. The fact that she was an insider enhanced her credibility, facilitated acceptance and access to the environment, created trusting relationships with potential participants, expedited the research process and facili- tated recruitment and collection of a rich store of data. This insider knowledge of the NICU allowed her to engage in discus- sions on an equal footing, and prevented misinterpretation of the information and facilitated the research process. However, to ensure participant perspectives were protected and researcher bias was limited, field notes and a researcher reflexivity journal was created to support continuous reflection by the researcher about her own values, preconceptions, behavior or presence during data collection thus enhancing the trustworthiness of the study findings (Jootun, 2009).

Data collection

Semi-structured interviews were conducted over a four-month period. Data collection was limited to core members of the interprofessional team. No parents were interviewed. An inter- view guide with open-ended questions was designed to explore participants’ perceptions of different facets of IPSDM. A total of 22 audiotaped interviews were completed: nurses (10), phy- sicians (5), respiratory therapists (3), and other health profes- sions (4). The majority of participants were female (96%), worked full time (77%), were very experienced NICU clinicians (73%), and worked both days and nights (64%) in their respec- tive roles in the NICU. The majority of interviews were com- pleted face-to-face (73%); however one dyadic interview was completed with two participants at their request. Four inter- views were carried out by phone. On average each interview lasted about 45 minutes (range, 30–90 minutes) with the dyadic interview taking the longest at 90 minutes in order to capture the contribution of both participants. A summary of partici- pants’ characteristics is provided in Table 1. Interviews were audiotaped with permission and transcribed verbatim.

Analysis

The audiotapes of the interviews were transcribed verbatim and entered into the NVivo 8® software program (QSR International, 2008). Data were anonymized to maintain con- fidentiality of the site, participants, and patients. A constant comparative method (Glaser & Strauss, 1967) was used during analysis and, in keeping with a qualitative approach, coding was driven by the data, using an inductive approach (Graneheim & Lundman, 2004). Initially, transcripts were read several times for a general sense of the content. The questions from the interview guide provided the initial orga- nizing framework for analysis. Individual verbatim responses were reviewed for similarities and recurring ideas, clustered

into specific themes and sub-themes, and grouped by profession.

Ethical considerations

Ethics approval for this study was received from the Research Ethics Boards at the participating hospital and the university.

Results

Participants discussed several features of IPSDM in NICU. Themes that emerged included: key participants and roles, components of the IPSDM process (collaboration, sharing, weighing, and building consensus), and outcomes (effects on decision quality and staff) of IPSDM. A summary of themes is presented below with example quotes for illustration. A visual representation of the relationships between the concepts is provided in Figure 1 see the online supplementary file.

Participants and key roles

Participants identified the following key roles important to IPSDM: (1) a leader who facilitates shared decision-making and, in some cases, takes responsibility for the decision; (2) professional experts who provide information and insight into the case; (3) someone who synthesizes and integrates the information together; (4) and the parents acting as surrogate decision makers, in relation to the role of the ‘synthesizer’ it was noted:

You are almost like a central computer - you get all this informa- tion coming at you, and try to make sense of it. Sometimes there are many ideas that are brilliant, but they are not feasible. At this point you try to find alternatives. This is very important for the benefit of the baby. I’m a connector. I’m the person that connects things. I try to make sense of all the inputs. It’s one of the most difficult things to do (MD1)

Although participants universally acknowledged the impor- tant role parents play and the need for their involvement in the decision-making process for preference-sensitive

Table 1. Participant characteristics.

Category Participant Characteristics (n = 22) %

Profession RN 10 45.5 MD 5 22.7 RT 3 13.6 OHP 4 18.2

Gender Male 1 4.5 Female 21 95.5

NICU Experience Very experienced (> 10 years) 16 72.7 Experienced (5–10 years) 5 22.8 Somewhat experienced (2–5 years) 1 4.5 Novice (< 2 years) 0 0

Work Rotation Days 6 27.3 Nights 2 9.1 Combination (days/nights) 14 63.6

Full/Part Time Status Full time 17 77.3 Part time 5 22.7

Interview Face-to-face (individual) 16 72.7 Phone (individual) 4 18.2 Group (1 group of 2 participants) 2 9.1

Code: RN (nurses), MD (physicians), RT (respiratory therapist), OHP (other health professionals).

286 S. I. DUNN ET AL.

decisions, there were diverse views about when and how the parents should be involved. Some participants indicated that the interprofessional team needed time to deliberate together first to review the situation and make sense of the data avail- able before meeting with the parents. Others indicated that the parent’s presence, even during these preliminary discus- sions, was appropriate:

When we meet with families, for complex cases for example, there is usually a different set of physicians, sometimes a dietician is involved, and often the bedside nurse or the primary care nurse, as well as, the parents, and we brainstorm together. Sometimes we inform the family of what our suggestion is [after the fact], but other times we just brainstorm together and come up with possibilities. (MD4)

The process of IPSDM

The following four key subthemes emerged from the data related to the process of IPSDM.

Collaboration Participants from all four professional groups agreed that IPSDM happens through a collaborative process of working together to identify the options to make a well-informed decision. Respondents used words such as ‘brainstorming, open discussion, and working as a team’ to illustrate this perspective. One of the nurses described this collaborative process as a joint effort to gather facts and deliberate about options to reach the best decision:

It’s a collaborative process, where everybody’s voice is listened to, and then there’s a joint decision made, on whatever issue has been addressed. (RN9)

Sharing Participants in this study interpreted sharing in three different ways: sharing information or professional expertise only, shar- ing during the deliberation/debate about treatment options, or sharing in making the decision itself. Respondents empha- sized the importance of sharing information or contributing professional expertise to the case as an essential part of the IPSDM process. There was general agreement across all pro- fessional groups about the importance of having as much information as possible, from as many perspectives as possi- ble, to make a well-informed decision in the best interests of the baby:

You’re sharing everybody’s knowledge, and everybody’s knowl- edge is coming from a different focus. Somebody might just have a different perspective all of a sudden that shines a different light on the situation that may be the solution. So you have to listen to all of that. (OHP1)

The second interpretation of sharing, highlighted by some of the nurses, other health professionals, and one physician, is about sharing in the deliberation or debate about options. From their perspective, IPSDM involves more than just shar- ing information, but it goes one step further, requiring the team to sift through all available information, deliberate, and identify options for consideration. The ultimate goal during these deliberations is to draw on the expertise within the team to come up with the options. However, for deliberation about

options to occur, information sharing within the team must happen first.

It’s a brainstorming type of session, where we offer what we can, what knowledge we can, what ideas, what suggestions, and what alternatives and you sift through that information and based on the patient, the clinical situation, and the best decision is made. Ultimately, the physician is going to be the one that has to make the final decision, but I feel that we are all contributing information, and gradually come to, some more quickly than others. (OHP1)

Most commonly, the interpretation of sharing involved not only sharing information, and sharing in the deliberation about options, but also sharing in the decision itself:

I would share my nursing expertise with the group and how my visions of what should be decided from my profession. Everyone gives their input and then together we make a decision on the care plan, or the issue that’s at hand. (RN7)

I think it’s very important to have a shared decision, because the technology has now greatly advanced, and so is the knowledge about a disease or about a machine that we can use. And I think culturally, and also professionally, and also physiologically speak- ing, sometimes the sharing of a decision makes you, pushes you, and challenges you to go further in your own reasoning, okay? (MD1)

Weighing According to participants, another important aspect of IPSDM involves weighing the options (pros and cons), weigh- ing the evidence and weighing the credibility of an opinion. Participants from all four professional groups spoke about weighing the options as a key step in the IPSDM process. Weighing the options involved having all the facts necessary to address each issue, sifting through the facts and synthesiz- ing the information, and brainstorming about the risk/bene- fits or pros/cons for each option:

Everyone had the opportunity to give some pointers and those pointers should include the risks and the benefits, the pros and cons and as many facts that you can have to make the best decision. (RN4)

All participants valued the contribution of experienced mem- bers of the team, but other forms of evidence were considered essential for decision-making too. These included: research evidence (scientific evidence); clinical evidence (current status of infant and family); professional experience (professional knowledge, tacit knowledge); practice-based evidence (what has worked before for self and others); and patient/family evidence (advocacy—defending patient/family rights)

The fact that participants were reflecting more about the evidence they need to consider as a team and advocating for the parents, rather than on how to engage patients or their families in decision-making, is consistent with practice where parents are not always included in the initial discussions until after the interprofessional team has had the opportunity to explore the issues first. Finally, participants spoke about the importance of weighing the credibility of the opinion put forward by those involved in the discussions. A number of factors influenced whether someone was perceived to be credible (e.g., experience, confidence, trust):

JOURNAL OF INTERPROFESSIONAL CARE 287

People do listen to me, because they trust me. It doesn’t mean they agree with me always, but they do trust me. Nurses need to learn how to present their case with evidence and confidence because that’s who gets listened to. (RN7)

I think that everybody’s opinion is very valid. I weigh very strongly some people’s opinion versus other people’s opinion. Some I may not have as much confidence in. (MD5)

Building Consensus Consensus was the most common method described by par- ticipants for reaching a decision. However, when the question of consensus was probed in this study, it became evident that consensus meant slightly different things to different people. To some participants, consensus meant achieving full agree- ment within the team or finding common ground through understanding and insight. To other participants, consensus meant the acceptance of another view, or simply agreeing to disagree, rather than being in full agreement.

Having consensus means everyone is in agreement. This is the right plan. (RN4)

I don’t think you have to agree. I think you have to be able to live with their decision. If they’ve given me good rationale as to why they’ve chosen something different from what I would choose, then I think I could live with the decision and feel I’d had a voice in the whole. (RN2)

A number of strategies were reported to facilitate achievement of consensus: providing input/exploring options, discussing and listening, respecting input provided by others, and under- standing/uncovering the meaning of best options:

It’s the act of people being listened to that a picture emerges. If we really sit down and talk, that something will emerge. I have been humbled because I’ve listened to somebody else who has thrown up a very different perspective of things. And then, because I’ve listened to what they’ve said, I thought they have a huge point here that I didn’t take into account quite like that. (RN9)

One of the respiratory therapists emphasized how important the diversity of opinion was to reaching consensus:

Every perspective about a decision to be made is there for a reason. It’s neither really right nor wrong. And we’ve arrived at that perspective because of our varied backgrounds, or varied experiences. And so there isn’t necessarily a right answer. But then, in keeping with the spirit of patient-centered, family-focused decision-making in a collaborative fashion, it may take some understanding of the different perspectives before you can reach consensus. So, the more you can uncover in an honest, open fashion, the more likely you’re going to reach consensus. (RT1)

A physician talked about the benefits of achieving consensus within the team. Not only does consensus make working together easier, it facilitates finding common ground, it rein- forces the plan, it decreases bias, and makes the provision of care easier for those who are responsible to carry out the plan.

If you have the consensus it means that you cannot be completely wrong. I think consensus is there to say, yes that was correct. It’s definitely easier to work in a field where people agree than where people disagree. It’s so difficult to do something because you have an order to do something, but you don’t agree with it? (MD2)

Barriers to Achieving Consensus: Despite the fact that consen- sus building has been identified as important for IPSDM,

achieving consensus within an interprofessional team (and with the parents) can be a challenge. Respondents highlighted three barriers: lack of information, differing values, and power differential within the interprofessional team. Lack of infor- mation can result because the common knowledge that exists between members of the interprofessional team or simply because people with something to say, who have valuable insight, just don’t speak up:

Because of the common knowledge, we’re not always in discus- sion and debate about the plan. (RN5)

Once we have discussed it fully, if people do not go out grum- bling, I believe that the team members are satisfied. One has to verbalize thoughts. (OHP3)

Respondents also perceived differing values as an obstacle to achieving consensus. A physician emphasized the importance of ‘doing no harm’ to guide decision-making to ensure the best interests of the patient are considered. Although the participants seem more focused on the barriers to working as a team in trying to achieve consensus rather than on the barriers to working with the parents, they did acknowledge differing values with the parents as a potential challenge as well. This is important given the fact that values clarification is an essential component of SDM.

We have to come to consensus. Maybe you’re not fully in agree- ment, but you have to look at the values, and where the parents are, because we are in a setting where parents are advocates for their child, their values might not be our values. Their standards might not be our standards, it’s to respect everyone. Ideally, it’s to respect everyone’s input. (OHP3)

Everybody has different values. You’re not going to come to a consensus necessarily on values. It comes down to the basic principles - do no harm. (MD5)

The final barrier to consensus decision-making identified by respondents from all four professional groups had to do with the power differential within the team and the fact that the physicians held the responsibility for the final decision. Participants’ perspectives seemed to be influenced by their expectations of who is involved and how decisions are actually made (e.g. consensus decision or decision made by the phy- sician following input from others), and how difficult con- sensus was to achieve:

I think when we’re sharing, we’re trying to get to a consensus of what the decision should be, but I think in the end if we don’t agree the physician will, if they’re strong on their hold, they allow everyone to have an opinion but they’ll ultimately make their decision. (RN7)

It depends whether it’s a treatment decision, or a medical approach decision, or whether it’s a social situation, or the well- being of the patient decision. I think the person involved in the decision is the one who really holds the responsibility on their shoulders? Legally the final responsibility is to the MD. (MD2)

As a way of countering the positional power differential that exists within the interprofessional team participants described a number of strategies, they used to ensure their perspectives were considered during IPSDM. These strategies included: being present for, being prepared, and participating in the discussion with confidence, tailoring their message to the

288 S. I. DUNN ET AL.

listener, directing the message to the person who is accoun- table and in authority, presenting a credible argument sup- ported by evidence, and being persistent.

You have to identify the problem, and then you have to have put in some thought as to how to solve it. It’s great to identify an unworkable situation or something that you think is going to be, is going to impact negatively. It’s really so much better to be able to identify that problem but also provide or suggest some solu- tions or pose some questions that may open some doors to some problem solving. (RN9)

Are you respected? Are you not respected? Are you presenting yourself in a manner that’s conducive to other people even listen- ing? Are you butting in? Are you suggesting? Are your opinions valid? Do you have sound background for bringing them up such that people can even entertain them as an option? If you’re just talking nonsense, then people aren’t going to listen. It’s all just in the demeanour; it’s all in the way that it’s brought to other people’s attention. (MD3)

Outcomes of IPSDM

The main outcome of IPSDM, identified by participants, was making a well-informed decision. Respondents indicated that a well-informed decision is a decision that takes into account all voices (including the parents), is based on empirical evi- dence and experience, and is made in the best interests of the infant after weighing pros and cons for each option. Well- informed decisions are better decisions than ones made with- out examining all the facts:

Shared decision-making is about all members of the team involved with the decision, having the opportunity to provide input in order to be able to make a well-informed decision. (RN4)

It’s important to discuss with other people then you know that you’re making the right decision. (MD3)

A secondary outcome of IPSDM identified by participants was that team members felt valued as participants in the decision- making process. Ultimately this resulted in increased morale in the group:

Shared decision making makes for better decisions. It’s an appre- ciation of people’s knowledge and experience. It makes each member feel valued. It’s so important to have different perspec- tives on decisions and it’s definitely contributed to morale. (RN5)

Discussion

This study aimed to explore and compare professional per- spectives about the key roles, processes, and outcomes of IPSDM in NICU. Health care providers described four key roles (professional expert, leader, synthesizer, and parent as an advocate for their baby and surrogate decision maker), key components of the IPSDM process (collaboration, sharing, weighing, and building consensus), and positive outcomes (effects on decision quality and morale) of IPSDM. Participants from all four professional groups agreed that IPSDM happens through a collaborative process of working together to make a well-informed decision. In general, per- spectives about the process of IPSDM across professional groups were similar with four notable exceptions: how differ- ent members of the interprofessional team interpret what

‘sharing’ means during SDM, the different types of evidence used by members of the interprofessional team, different perspectives on when parents should be engaged in the pro- cess, and perspectives about power and control as a barrier to IPSDM. The findings discussed below increase our under- standing of the roles and processes involved in IPSDM in NICU and provide insight to further inform the IP-SDM model and IPC.

Inclusion of professional experts and parents or surrogate decision makers in the process of decision-making has been acknowledged in other studies (Curtis & Tonelli, 2011; Légaré et al., 2011a), as is the role of leader to initiate and support the SDM process (Kryworuchko et al., 2012). The synthesizer role, however, was a new role identified in this study as key to IPSDM.

The ‘leader role’ described by participants in this study is different from the ‘decision coach’ role described in the IP- SDM Model which defines decision coach as a trained HCP who provides non-directional individualized support to patient’s and families involved in healthcare decision-making, but who does not make the decision for the patient (Stacey, Murray, Dunn, Menard, & O’Connor, 2008). The leader role in our study is also somewhat different from that described by Kryworuchko et al. (2012) – where the role was to initiate the decision process, identify the options, and make the decision. The participants in our study highlighted the need for a leader experienced in the process of SDM to shepherd the team through the process of information exchange, sharing, delib- eration and consensus building to identify the decision to be made and the options for consideration. It is not sufficient that individual HCPs contribute their expertise if they do not understand or have experience with the process of SDM. The leader leads the SDM process like the conductor of an orches- tra; whereas a decision coach would focus attention on ensur- ing parents have the opportunity and skills to participate in the process.

The synthesizer role was highlighted as a specific skill set of the physicians by participants in this study. The synthesizer requires the expertise to not only guide the discussion, but also to critically appraise, interpret and integrate large amounts of diverse information as the interprofessional team deliberates about the evidence and the pros and cons of the options to facilitate consensus building within the group.

With respect to the parents’ role in IPSDM, all participants in this study acknowledged the important role parents play in decision-making, but differed in their view about when par- ents should be engaged in the IPSDM process. This is con- sistent with the view that family participation in decision making is a cornerstone of patient- and family-centered care (Boland et al., 2016), and the research that demonstrates parents want to be involved in decisions about their infants, but vary in their desired level of involvement and when and how they become involved (Payot, Gendron, Lefebvre, & Doucet, 2007).

For some HCPs timing of parent involvement was driven more by parents’ proximity and availability—an issue that presents a logistical challenge in NICU where parents and babies are often separated. However, some nurses and physi- cians spoke about the need for the interprofessional

JOURNAL OF INTERPROFESSIONAL CARE 289

healthcare team to discuss the case first. The rationale for this phased approach was to allow time for the interprofessional team (and any external HCPs involved in the infant’s care) to explore all the facts, fill in the gaps, and prepare for discussion with the parents in order to present consistent and complete information and clearly articulate the decision to be made and the options for consideration. The need for regular debriefing meetings between physicians and nurses have been acknowl- edged elsewhere and are associated with greater satisfaction for families in ICU (Azoulay et al., 2014).

The phased approach to parent engagement is also consis- tent with the early stages of the IP-SDM model where the role of the HCP is to initiate the process of SDM and facilitate information exchange about the decision to be made, the options to be considered and the family’s values and prefer- ences (Légaré et al., 2011a). However, the model does not stipulate that ‘the HCP who initiates’ the SDM process should present the interprofessional team consensus view about a decision to be made or the options—or just present their own view. Our data would suggest that in an NICU environ- ment where patient acuity is high, cases are complex and many services are involved in providing care, the phased approach to engaging parents in SDM is appropriate and would ensure the ‘initiator’ has consulted with other members of the interprofessional team and has more complete informa- tion to share before meeting with the family.

There are, however, potential risks of the phased approach. First, since diverse professional perspectives can be a barrier to initiating the SDM process, if the team cannot reach con- sensus about the clinical situation and the need for a decision and the options for consideration, the process can be stalled. Second, delayed engagement with parents increases the risk that important information about the family will not be con- sidered during the preliminary discussions (Kryworuchko et al., 2012). Therefore, it is important to ensure the team has time to clarify and understand the clinical situation, the parents’ values and preferences are elicited, and parents are engaged in the IPSDM process at the level they desire (Azoulay et al., 2014; Kryworuchko et al., 2012).

The concept of sharing, which has been identified as a key component of interprofessional collaborative practice (Canadian Interprofessional Health Collaborative, 2010; D’Amour et al., 2005; Reeves et al., 2017), was also identified by participants in this study as an essential component of IPSDM. However, sharing meant different things to different people: sharing information, sharing in deliberations, or shar- ing in the decision and differences were noted across profes- sional groups.

Most often, participants from the OHP group referred to the first level of involvement when speaking about SDM— contributing their expertise when consulted. In contrast, nurses and respiratory therapists often described SDM as a process where everyone provides input and following discus- sion a consensus decision is made collectively—with an expec- tation that there would be ‘no decision without me’. Physicians spoke about SDM involving all three levels of involvement as well - gathering input from experts (including the family), and discussing the options together so a decision could be made. Physicians acknowledged themselves and were

acknowledged by all other HCP groups to be responsible for the decision, but they did not describe themselves as the ultimate decision maker—whereas nurses and respiratory therapists did acknowledge the positional power they held on the team.

The different interpretations offered by participants during interviews is consistent with the results of an earlier survey done for this study (Dunn et al., 2013), which found that HCPs views differ about what constitutes optimum IPSDM. This suggests that inconsistencies between expectations and the reality of IPSDM can create confusion and dissatisfaction with the process if consensus cannot be reached. This view is also aligned with the four-step decision process outlined by DeKeyser Ganz, Engelberg, Torres, and Curtis (2016) which includes different levels of information exchange, deliberation and sharing as a precursor to the final decision being made by the multi-level team.

The differing views about what ‘sharing’ means in SDM and when it happens not only varied across professional groups but also within groups suggesting that the professional lens through which different HCPs see the clinical world is not the same. This may be influenced by their expectations, previous experiences, their roles within the unit and their proximity to the actual bedside care of the patient (e.g. arms-length involvement versus direct hands-on care over prolonged time periods). These differing views have signifi- cant implications for practice and education and highlight the importance of establishing ‘rules of engagement’ for IPSDM so all members of the interprofessional team know what to expect, how to participate, and how decisions will be made when differing professional perspectives about a case exist.

The concept of weighing the evidence and weighing the pros and cons of each option identified by participants in this study is consistent with studies about IPSDM in NICU (Kavanaugh, Savage, Kilpatrick, Kimura, & Hershberger, 2005) and the IP-SDM model (Légaré et al., 2011a). However, participants in this study also acknowledged the need to weigh the credibility of an opinion during IPSDM. This idea is consistent with key features of successful IPC that indicates HCPs are more likely to collaborate with clinicians they respect, trust and perceive to have pertinent knowledge, are more experienced and therefore more competent (Gregory & Austin, 2016; Pullon, 2008).

Although all participants valued different forms of evi- dence and considered the contribution of experienced, cred- ible members of the team, as essential for decision-making, the types of evidence commonly used by different HCPs varied. Physicians frequently referred to quantitative research evidence when weighing the options for care, while nurses and respiratory therapists referred to their experience and practice-based evidence more often. The consideration of family perspectives, values, and preferences, either through direct input or advocacy, was acknowledged as important for preference-sensitive decisions by all.

All participants in the current study acknowledged the importance and value of a consensus decision—a view that is consistent with current guidelines for ethical decision-mak- ing in NICU (Baumann-Holzle, Maffezzoni, & Bucher, 2005) and IPC (Canadian Interprofessional Health Collaborative,

290 S. I. DUNN ET AL.

2010). Physicians talked about the benefits of achieving con- sensus within the team (e.g. makes working together easier, facilitates finding common ground, reinforces the plan, decreases bias, and makes the provision of care easier for those who are responsible to carry out the plan), with a positive view that if you have consensus it means you cannot be completely wrong. While nurses talked about agreeing to disagree if a decision was made that differed from their view, but the rationale provided by someone else was convincing enough.

Participants also identified the power differential that exists within the interprofessional team as one of the most challen- ging barriers to achieving consensus (Coombs, 2003; Coombs & Ersser, 2004). Power, which is the capability of one party to exert influence on another to act in a prescribed manner (Panteli & Tucker, 2009), can come from different sources: a) referent power (delegated authority or positional power), b) legitimate power (social position or professional status), c) expert power (knowledge and expertise), d) informational power (persuasion), e) coercive power and, f) reward power (Raven, 1993).

Interprofessional collaboration and shared decision-mak- ing imply participants are equal partners in decision-making (Rose, 2011). However, in reality, members of the IP team in an NICU are not all equal. Someone is ultimately responsible for the decision and, as pointed out by participants, this is often the physician. Power differentials also exist because of knowledge, experience, and role diversity within an interpro- fessional team placing the holders of knowledge in a position of power in the decision-making process (Coombs, 2003; Coombs & Ersser, 2004). The power differential most com- monly referred to by participants in this study stemmed from the positional authority physicians hold in NICU.

This is consistent with other studies that have demon- strated decision-making in ICU settings continues to be strongly driven by the medical knowledge base and authority and other sources of knowledge and roles, such as those held by nurses, may be less valued by some physicians resulting in tension between nursing and medicine (Coombs, 2003; Coombs & Ersser, 2004; Reeves et al., 2015). In this study the interprofessional team was experienced, highly function- ing, and considered IPSDM a normal part of their collabora- tive practice. There was a general expectation that decisions were based on consensus. However, if decisions were made without consensus, they were more likely to be aligned with the physician’s perspective resulting in frustration for the other members of the team.

Since professionals in a position of power can influence decision-making and the treatment plan (Joseph-Williams et al., 2014; Rose, 2011), it is essential that practice models address the power disparities which are a reality in any inter- professional team to ensure equitable participation in the process of decision-making is possible. Attempts to deal with this issue by focusing on the interpersonal development of nurses rather than challenging the dominant role of med- icine have proven ineffective (Coombs, 2003). Whereas, valu- ing and sharing knowledge about the patient in a process of trade has been reported to facilitate not only the exchange of

information, but also an exchange of power as team members negotiate with one another (Lingard, Espin, Evans, & Hawryluck, 2004). Although recommended as a potential strategy, in reality, this does not often happen in practice (Fox & Reeves, 2015).

However, the actual exchange of information within the team in a process of trade may not be sufficient to ensure consensus is achieved. Nurses, respiratory therapists, and other health professionals described a number of strategies they used to be persuasive and ensured their opinions were heard during IPSDM. These strategies included: presenting solutions, backing up their opinion with evidence, speaking with confidence, being persistent, and directing the message to the person who is accountable and in authority. Being a persuasive messenger provided these HCPs with a source of informational power to balance the positional power disparity within the team. Physicians, on the other hand, spoke less about needing to use persuasion themselves to get their mes- sage across, and more about what other members of the interprofessional team needed to do to be ‘heard’.

This research has implications for interprofessional educa- tion (IPE), practice, parent engagement, and research. The findings suggest that IPE programs, aimed at strengthening HCPs skills and confidence in participating in IPSDM as equals, would facilitate this process (Canadian Interprofessional Health Collaborative, 2010). The goals of IPE should not only focus on role clarification, team function- ing, collaborative leadership, IPC and conflict resolution, but also enhance the abilities of different HCPs to be leaders to facilitate the IPSDM process, and to synthesize information from different sources.

There are a number of implications for practice in NICU. First, our results highlight the need to establish clear policies, procedures, and ‘rules of engagement’ to guide clinicians through the SDM process and ensure all members of the inter- professional team know what to expect, how to participate, and how decisions will be made when differing professional perspec- tives about a case exist. Although different perspectives were evident across professional groups in this study, this may have been the result of other factors (e.g., previous experience or level of confidence), and not just due to their professional affiliation. Given this fact, it is important to not assume that everyone within a particular professional group thinks alike, has the same level of skill, and interprets the meaning of SDM and their role in the process the same way. Clarification of the SDM process will help to align HCP expectations with the reality of practice, reduce frustration, and increase satisfaction with the process of IPSDM for complex cases.

Secondly, it is important to create opportunities for all members of the interprofessional team to interact together (either face-to-face or virtually) prior to meeting with the family, so they can sift through the facts of the case together, synthesize information and explore all the options. This will help to prevent decisions based on incomplete information, and ensure parents receive a comprehensive interprofessional team view of the decision to be made and options for con- sideration to facilitate their involvement in the actual IPSDM process.

JOURNAL OF INTERPROFESSIONAL CARE 291

There are also important implications related to parent involvement in IPSDM based on the findings of this study. First, it is important to establish clear and transparent pro- cesses to engage with parents soon after admission to explore their preferences around the level of involvement in decision- making and identify strategies to ensure their input is elicited. This is particularly important for situations when parents are separated from their infant, may not be able to visit together, and may not be present when critical meetings of the full interprofessional team are held. In addition, it is important to not only provide support to help parents through the deci- sion-making process, but also to increase their confidence to speak up when together with the full interprofessional team.

Finally, there are implications for further refinement of the IP-SDM model. The model, as currently designed, presents the IPSDM process as very linear, with information exchange occurring after the ‘decision to be made’ has been identified. It does not address the complexities of decision-making within an interprofessional team in NICU when multiple HCPs are involved (both within the core team and external to that team), perspectives about the clinical situation differ, patient problems are complex, consensus is difficult to achieve, parents presence may not always be possible when the interprofessional team is available to meet, and parents’ support needs vary. The question remains—is it appropriate for any HCP to ‘initiate the process’ before the team has had sufficient opportunity to explore the clinical situation to determine whether a preference-sensitive decision needs to be made (e.g., decision about withdrawal or withholding of care), and what options are to be considered? The results of this study suggest that the entry point into the model may need to be flexible for ICU settings, to allow information exchange among the interprofessional team to happen prior to initiating the IPSDM process. Parents may be involved in these preliminary discussions but this should be determined based on the parents’ preferences, their availability, and the complexity of the patient care issues.

Further research is needed to explore the nature of IPSDM with different interprofessional teams, in a variety of settings, and in situations where HCPs cannot meet face- to-face. Research is also needed to identify the most effec- tive strategies to enhance the knowledge and skills of all members of the interprofessional team, so they can partici- pate fully in the process of IPSDM and develop the specific competencies required for the leadership and synthesizer roles. In addition, research about the barriers to IPSDM in intensive care would help to identify strategies and develop resources to effectively target profession-specific issues related to this process. Finally, further research is needed to increase understanding about the barriers to parents involvement in IPSDM, and whether the addition of a decision coaching role to the interprofessional team as identified by Stacey et al. (2008), would facilitate parent involvement in the process of IPSDM in NICU.

There are a number of potential limitations to this study: recall bias, social desirability bias, and the fact this was a small study involving a single site. A number of strategies were used

to ensure the trustworthiness of the results. This study explored IPSDM in-depth through the experiences and per- ceptions of the interprofessional team working in one NICU, which limits transferability of the findings to other sites. Although the sample group was limited to those practitioners working in the unit at the time of data collection, multiple perspectives were obtained across all professional groups making up the interprofessional team and a rich source of data was collected during the inquiry. Participants recruited for this study were all familiar with the interprofessional model of practice in this unit, and an interview guide was used to probe the professional perspectives about the IPSDM process. Clustering participants from different professions together in the OHP group to ensure anonymity may have masked some of the potential heterogeneity of responses from those individuals. However, participant data from all groups (no matter the professional affiliation) was coded separately to allow for individual variation to be acknowledged. The influ- ence of social desirability bias was limited through voluntary participation, establishing a trusting relationship with partici- pants, ensuring confidentiality of responses, and reporting only anonymized results.

The processes used for data collection were simple, trans- parent and reproducible. Data triangulation (different health- care professionals’ perspectives about similar events), increased the dependability of the results. Confirmability was established by verifying processes and findings through purposive and maximum variation sampling, and investigator responsiveness to participants throughout the interview and observation pro- cess. These facts give our findings substantial weight.

In summary, the perspectives of HCPs involved in decision- making in intensive care are important for the quality of the decisions made. Information gleaned through key informant interviews indicated that IPSDM requires HCPs to have knowl- edge, skills, and confidence to participate fully, as well as access to protocols to facilitate the decision-making process when opinions differ about the best way forward.

Declaration of interest

The authors report no conflict of interests. The authors alone are responsible for the writing and content of this article.

References

Azoulay, E., Chaize, M., & Kentish-Barnes, N. (2014). Involvement of ICU families in decisions: Fine-tuning the partnership. Annals of Intensive Care, 4(37), 1–10.

Baggs, J. G., Schmitt, M. H., Mushlin, A. I., Mitchell, P. H., Eldredge, D. H., Oakes, D., & Hutson, A.D. (1999). Association between nurse- physician collaboration and patient outcomes in three intensive care units. Critical Care Medicine, 27(9), 1991–1998.

Barry, M., & Edgman-Levitan, P. (2012). Shared decision making - The pinnacle of patient-centred care. New England Journal of Medicine, 366(9), 780–781.

Baumann-Holzle, R., Maffezzoni, M., & Bucher, H. U. (2005). A frame- work for ethical decision making in neonatal intensive care. Acta Pediatrica, 94, 1777–1783.

292 S. I. DUNN ET AL.

Boland, L., McIsaac, D. I., & Lawson, M. (2016). Barriers to and facil- itators of implementing shared decision making and decision support in a pediatric hospital: A descriptive study. Paediatrics Child Health, 21(3), e17–E21.

Canadian Interprofessional Health Collaborative. (2010). A national interprofessional competency framework. Vancouver, BC: Author.

Charles, C., Gafni, A., & Whelan, T. (1999). Decision-making in the physician-patient encounter: Revisiting the shared treatment decision- making model. Social Science & Medicine, 46, 651–661.

Coombs, M. (2003). Power and conflict in intensive care clinical decision making. Intensive and Critical Care Nursing, 19, 125–135.

Coombs, M., & Ersser, S. J. (2004). Medical hegemony in decision- making – A barrier to interdisciplinary working in intensive care? Journal of Advanced Nursing, 46(3), 245–252.

Curtis, J., & Tonelli, M. (2011). Shared-decision making in the ICU; Value, challenges, and limitations. American Journal of Respiratory Critical Care Medicine, 183, 840–841.

D’Amour, D., Ferrada-Videla, M., Rodriguez, L. S., & Beaulieu, M. (2005). The conceptual basis for interprofessional collaboration: Core concepts and theoretical frameworks. Journal of Interprofessional Care, 19(Supplement 1), 116–131.

DeKeyser Ganz, F., Engelberg, R., Torres, N., & Curtis, J. (2016). Development of a model of interprofessional shared clinical decision making in the ICU: A mixed methods study. Critical Care Medicine, 44(4), 680–689.

Dunn, S., Cragg, E., Graham, I. D., Medves, J., & Gaboury, I. (2013). Interprofessional shared decision making in NICU: A survey of the interprofessional healthcare team. Journal of Research in Interprofessional Practice and Education, 3(1), 63–77.

Elwyn, G., Edwards, A., Kinnersley, P., & Grol, R. (2000). Shared deci- sion making and the concept of equipoise: The competencies of involving patients in healthcare choices. British Journal General Practice, 50(460), 892–899.

Elwyn, G., Frosch, D., & Rollnick, S. (2009). Dual equipoise shared decision making: Definitions for decision and behaviour support interventions. Implementation Science, 4(75), 1–8.

Fox, A., & Reeves, S. (2015). Interprofessional collaborative patient- centred care: A critical exploration of two related discourse. Journal of Interprofessional Care, 29(2), 113–118.

Gachoud, D., Albert, M., Kuper, A., Stroud, L., & Reeves, S. (2012). Meanings and perceptions of patient-centeredness in social work, nursing and medicine: A comparative study. Journal of Interprofessional Care, 26, 484–490.

Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research. Chicago, US: Aldine de Gruyter.

Graneheim, U., & Lundman, B. (2004). Qualitative content analysis in nursing research: Concepts, procedures and measures to achieve trustworthiness. Nurse Education Today, 24(2), 105–112.

Gravel, K., Légaré, F., & Graham, I. D. (2006). Barriers and facilitators to implementing shared decision-making in clinical practice: A systema- tic review of health professional’s perceptions. Implementation Science, 1(16), 1–12.

Gregory, P., & Austin, Z. (2016). Trust in interprofessional collaboration: Persepctives of pharmacists and physicians. Canadian Pharmacy Journal, 149(4), 236–245.

Grinspun, D. (2007). Healthy workplaces: The case for shared clinical decision making and increased full-time employment. Healthcare Papers, 7, 85–91.

Haggerty, J. L., Reid, R. J., Freeman, G. K., Starfield, B. H., Adair, C. E., & McKendry, R. (2003). Continuity of care: A multidisciplinary review. British Medical Journal, 327(7425), 1219–1221.

Jain, M., Miler, L., & Belt, D. (2006). Decline in ICU adverse events, noscomial infections and cost through a quality improvement initia- tive focusing on teamwork and culture change. Quality and Safety in Health Care, 15, 235–239.

Jootun, D. (2009). Reflexivity: Promoting rigour in qualitative research. Nursing Standard, 23(23), 42–46.

Joseph-Williams, N., Elwyn, G., & Edwards, A. (2014). Knowledge is not power for patients: A systematic review and thematic synthesis of

patient-reported barriers and facilitators to shared decision making. Patient Education and Counseling, 94, 291–309.

Joseph-Williams, N., Lloyd, A., Edwards, A., Stobbart, L., Tomson, D., Macphail, S., Dodd, C.,Brain, K.,Elwyn, G., & Thomson, R. (2017). Implementing shared decision making in the NHS: Lessons from the MAGIC programme. British Medical Journal, 357, 1–6.

Kavanaugh, K., Savage, T., Kilpatrick, S., Kimura, R., & Hershberger, P. (2005). Life support decisions for extremely premature infants: Report of a pilot study. Journal of Pediatric Nursing, 20(5), 347–359.

Kon, A., Davidson, J., Morrison, W., Danis, M., & White, D. (2016). Shared decision making in intensive care units: An American College of Critical Care Medicine and American Thoracic Society Policy Statement. Critical Care Medicine, 44(1), 188–201.

Kryworuchko, J., Stacey, D., Peterson, W., Heyland, D., & Graham, I. (2012). A qualitative study of family involvement in decisions about life support in the Intensive Care Unit. American Journal of Hospice & Palliative Medicine, 29(1), 36–46.

Lambert, V., & Lambert, C. (2012). Qualitative descriptive research: An acceptable design. Pacific Rim International Journal of Nursing Research, 16(4), 255–256.

Légaré, F., Ratte, S., Gravel, K., & Graham, I. D. (2008). Barriers and facilitators to implementing shared decision-making in clinical prac- tice; Update of a systematic review of health professionals’ percep- tions. Patient Education and Counseling, 73, 526–535.

Légaré, F., Stacey, D., Gagnon, S., Dunn, S., Pluye, P.,Frosch, D., Kryworuchko, J.,Elwyn, G.,Gagnon, M. P., &Graham, I. D. (2011a). Validating a conceptual model for an interprofessional approach to shared decision-making: A mixed methods study. Journal of Evaluation in Clinical Practice, 17(4), 554–564.

Légaré, F., Stacey, D., Pouliot, S., Gauvin, F. P., Desroches, S., Kryworuchko, J., Dunn, S.,Elwyn, G.,Frosch, D., Gagnon, M.P., Harrison, M.,Pluye, P.&,Graham, I. D. (2011b). Interprofessionalism and shared decision-making in primary care: A stepwise approach towards a new model. Journal of Interprofessional Care, 25, 18–25.

Lingard, L., Espin, S., Evans, C., & Hawryluck, L. (2004). The rules of the game: Interprofessional collaboration on the intensive care unit team. Critical Care, 8(6), 403–408.

Makoul, G., & Clayman, M. L. (2006). An integrative model of shared decision-making in the medical encounter. Patient Education & Counseling, 60, 301–312.

McCleary, L. (2002). Equipoise in clinical nursing research. Canadian Journal of Nursing Research, 34(3), 49–60.

Palinkas, L., Horwitz, S., Green, C., Wisdom, J., Duan, N., & Hoagwood, K. (2016). Purposeful sampling for qualitative data collection and anlysis in mixed method implementation research. Admin Policy Mental Health, 42(5), 533–544.

Panteli, N., & Tucker, R. (2009). Power and trust in global virtual teams. Communications of the ACM, 52(12), 113–115.

Payot, A., Gendron, S., Lefebvre, F., & Doucet, H. (2007). Deciding to resuscitate extremely premature babies: How do parents and neona- tologists engage in the decision? Social Science & Medicine, 64(2007), 1487–1500.

Pullon, S. (2008). Competence, respect and trust: Key features of success- ful interprofessional nurse-doctor relationships. Journal of Interprofession Care, 22(2), 133–147.

Puntillo, K., & McAdam, A. (2006). Communication between physicians and nurses as a target for improving end-of-life care in the intensive care unit: Challenges and opportunities for moving forward. Critical Care Medicine, 34, S332–S340.

QSR International. (2008). NVivo qualitative data analysis Software (Version 8).Australia: QSR International Pty Ltd. Australia: Author.

Raven, B. H. (1993). The bases of power: Origins and recent develop- ments. Journal of Social Issues, 49(4), 227–251.

Reader, T., Flin, R., & Lauche, K. (2006). Non-technical skills in the intensive care unit. British Journal of Anaesthesia, 96, 551–559.

Reeves, S., McMillan, S. E., Kachan, N., Paradis, E., Leslie, M., & Kitto, S. (2015). Interprofessional collaboration and family member involve- ment in intensive care units: Emerging themes from a multi-sited ethnography. Journal of Interprofessional Care, 29(3), 230–237.

JOURNAL OF INTERPROFESSIONAL CARE 293

Reeves, S., Pelone, F., Harrison, R., Goldman, J., & Zwarenstein, M. (2017). Interprofessional collaboration to improve professional prac- tice and healthcare outcomes. Cochrane Database Systematic Review, Review, CD000072. doi:10.1002/14651858.CD000072.pub3

Rose, L. (2011). Interprofessional collaboration in the ICU: How to define? Nursing in Critical Care, 16(1), 5–10.

San Martin-Rodriguez, L., Beaulieu, M. D., D’Amour, D., & Ferrada- Videla, M. (2005). The determinants of successful collaboration: A review of the theoretical and empirical studies. Journal of Interprofessional Care, 19(Suppl 1), 132–147.

Sandelowski, M. (2010). What’s in a name? Qualitative description revis- ited. Research in Nursing & Health, 23, 77–84.

Shay, L., & Lafata, J. (2017). Where is the evidence? A systematic review of shared decision making and patient outcomes. Medical Decision Making, 35(1), 114–131.

Stacey, D., Murray, M. A., Dunn, S., Menard, P., & O’Connor, A. M. (2008). Decision coaching to support shared decision making: A framework, evidence and implications for nursing practice, education, and policy. Worldviews on Evidence-Based Nursing, 5(1), 25–35.

Van den Bulcke, B., Vyt, A., Vanheule, S., Hoste, E., Decruyenaere, J., & Benoit, D. (2016). The perceived quality of interprofessional team- work in an intensive care unit: A single centre intervention study. Journal of Interprofessional Care, 30(3), 301–308.

Visser, M., Deliens, L., & Houttekier, D. (2014). Physician-related bar- riers to communication and patient- and family-centred decision- making towards end of life in intensive care: A systematic review. Critical Care, 18(604), 1–19.

Vivian, L., Marais, A., McLaughlin, S., Falkenstein, S., & Argent, A. (2009). Relationships, trust, decision-making and quality of care in a pediatric intensive care unit. Intensive Care Medicine, 35, 1593–1598.

294 S. I. DUNN ET AL.

Copyright of Journal of Interprofessional Care is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.