Making it real: the institutionalization of collaboration through formal structure

ABSTRACT

Collaboration has achieved widespread acceptance as an indispensable element of healthcare delivery in recent decades, despite modest evidence for its impact on healthcare outcomes. Attempts to understand this seeming paradox have been based mostly in functionalist or conflict-theoretical approaches. Currently lacking, however, is an articulation of how collaborative ideals are embedded in broadly shared beliefs about what healthcare is and how it operates. In this article, we examine how language used in the CanMEDS competency framework and in two guides for Family Health Teams construct idealized versions of rational, autonomous physicians and primary care organizations, respectively. Informed by phenomenological sociology and neo-institutional theory, we characterize these documents as elements of formal structure, the putative “blueprints” for healthcare planning and activity. Drawing on this analysis, we argue that these documents and “collaborative” formal structures in general, not only function as tools to make healthcare more collaborative, but also create an appearance of “real” collaboration, independently of the realities of practice. We argue that they thus instill confidence that the current healthcare system functions according to deep-seated societal values of justice and progress. We conclude by emphasizing the potentially distorting influence of this on efforts to understand and improve healthcare.

KEYWORDS:

• Organizational studies
• interprofessional collaboration
• institutional analysis
• health services research
• phenomenology
• interagency

Introduction

The last fifty years have seen a substantial increase in interest, study and investment in collaboration as a model of healthcare planning and delivery (Paradis & Reeves, 2013). Collaborative care emphasizes and valorizes an idealized image of healthcare; one delivered at the micro level by teams of clinicians from different occupations, and at the macro level by integrated systems composed of collaborative healthcare organizations. This “collaborative ideal” acknowledges and pays tribute to the multiplicity of perspectives and contributions that constitute contemporary healthcare and endorses collaborative practice to improve the quality of care delivery. Indeed, collaboration has been proposed as a solution to a wide array of problems, from ineffective care to treatment errors, a fragmented system, patient and provider dissatisfaction and resource waste (Paradis & Whitehead, 2018). This is despite the fact that efforts to make care processes more collaborative have resulted in rather modest impacts on care quality and health outcomes (Reeves, Pelone, Harrison, Goldman, & Zwarenstein, 2017; Reeves, Perrier, Goldman, Freeth, & Zwarenstein, 2013).

Researchers have pursued two broad avenues of inquiry to understand collaborative care. The first, a functionalist approach, focuses on the causal factors influencing collaborative behaviors and the relative success of interventions aimed at fostering them (Marlow, Lacerenza, Paoletti, Burke, & Salas, 2018; Neily et al., 2010). The primary goal of this research is to improve care systems’ functionality to optimize desired outputs (e.g. patient health, patient experience) while minimizing cost and error. A second avenue of inquiry, based on conflict theories, examines the differential construction and enactment of collaboration by different groups, particularly the different professions (Baker, Egan-Lee, Martimianakis, & Reeves, 2011; Bell, Michalec, & Arenson, 2014; Haddara & Lingard, 2013; Kuper & Whitehead, 2012; Whitehead, 2007). Authors working in this tradition tend to critically problematize collaboration, focusing on the divergent interests, practices, systems of knowledge and discourses that produce, reinforce or challenge interprofessional hierarchies.

A third theoretical approach has had little uptake to date but offers great potential in understanding the gap between the rhetoric and outcomes of collaborative care: a neo-institutional perspective on collaboration (Paradis & Reeves, 2013; Suter et al., 2013). This approach, rooted in a (macro) phenomenological paradigm (Powell & DiMaggio, 1991), can help reveal how the underlying values and premises of the collaborative care movement are transformed into basic assumptions of what healthcare is, and how it is delivered. A neo-institutional perspective on collaboration can shed light on how the ideal of collaborative care is transformed into the taken-for-granted structures of practice. It may also help to explain why, despite seemingly ubiquitous enthusiasm (see Paradis & Whitehead, 2018), evidence of impact on desired outcomes is, to date, so modest. By focusing on the normative aspects of collaboration – rather than how it works or on how power defines it – and the creation of structures that reflect widely-shared cultural norms, this perspective highlights the powerful role of ideas and values in influencing individual and organizational behavior.

In this article, we examine how the collaborative ideal is expressed in documents developed and used with the stated purpose of making the routine, everyday reality of healthcare more collaborative. We examine how ideas and values basic to modern western culture are mobilized within collaborative “formal structures,” the putative “blueprint[s]” (Meyer & Rowan, 1977, p. 342) for organizational activities and functions. In other words, we examine how the language of collaboration, mobilized within these “blueprints,” makes collaboration “real”.

Our theoretically informed argument, anchored in documentary analysis, suggests that in expressing collaborative norms, these documents or “formal structures” officially align the organizations and professionals they purport to depict with an idealized vision of modern society. This ideal posits a society in which effective, efficient, coordinated activity occurs with minimal coercion, through cooperation between autonomous actors (Meyer, Boli, Thomas, & Ramirez, 1997; Meyer & Bromley, 2013). The language of these documents constructs idealized clinicians and organizations who are autonomous but bounded, constrained to act appropriately by their very nature rather than by coercive forces (Meyer & Bromley, 2013).

Toward a sociological, phenomenological understanding of collaboration

The principal aim of this analysis is to show how an idealized picture of modern society has been embedded within some formal structures of healthcare, the organizational “blueprint[s] for activities” (Meyer & Rowan, 1977, p. 342), intended to foster collaborative care. In doing so, we apply a theoretical framework informed by phenomenological sociology (Berger & Luckmann, 1966) and, more particularly, one strand of it: the Stanford school of neo-institutional theory (Bromley & Powell, 2012; DiMaggio & Powell, 1983; Meyer et al., 1997; Meyer & Bromley, 2013; Meyer & Rowan, 1977; Paradis & Reeves, 2013; Powell & DiMaggio, 1991).

Sociological phenomenology borrows concepts from philosophical phenomenology, an approach to theory of knowledge (i.e. epistemology) developed by Edmund Husserl. As conceived by Husserl, phenomenology is a method for describing the basic character and structure of the perceived world from the first-person perspective; the world that is “given” to the perceiving subject, rather than the world that exists apart from them, outside of their perception (i.e. the objective world). The phenomenological world consists of “things” and structures defined by their meaning or significance for the perceiving subject, things to be observed, picked up, thrown, opened, considered, referred to, described, evaluated, improved, worsened, etc. Although these “things” are laden with meaning, they are taken for granted as elements of the world of the perceiving subject (Smith, 2018).

Sociologist Alfred Schutz and, subsequently, Berger and Luckman adapted ideas from philosophical phenomenology to the analysis of the social world, giving rise to phenomenological sociology. Instead of focusing on the world “given” to the single perceiving subject, they focused on the world that is given, or taken for granted, in the routine social life of groups, institutions and societies (Berger & Luckmann, 1966). So, for example, individual perceiving subjects will generally see someone with a white coat and a stethoscope as a physician, someone trustworthy with whom to discuss health problems. For some members of marginalized groups, however, this physician image might invite mistrust, and therefore lead to a different kind of interaction with the healthcare system (Nicolaidis et al., 2010; Peek et al., 2010).

For Berger and Luckmann (1966), the given or routinized, taken-for-granted reality of everyday life consists in the individual world of the perceiving subject, the intersubjective world of human interaction, and the objective world common and accessible to all. Through “internalization”, experiences with the external world and human interactions socialize individuals. Through “externalization”, shared meanings are negotiated between individuals through communication and interaction. Through “objectivation”, ideas and norms are manifested in the objective, enduring objects and structures that make up the social world or environment. In other words, they become part of the world we can all observe, act upon, and be acted upon by. This is not to say that neo-institutionalists believe that there is only one world out there – that they believe in a fully objective science – but rather that they emphasize the power of norms as they are reified in documents and organizational forms.

When objects and structures endure for a long time, we come to see them and their associated ideas and norms as permanent, inevitable, or natural, rather than as historically contingent, changeable and human-made (Berger & Luckmann, 1966). Ideas and normative preferences that are subjective, contestable and changeable are consequently transformed into the objects and structures that make up the taken-for-granted “reality” of routine social life. So, for example, we may come to take for granted that healthcare is provided by “clinicians” rather than, say, unique individuals, and by “healthcare organizations” rather than by a variety of more or less stable affiliations.

Building on these ideas, Meyer and Rowan (1977) produced a seminal article defining the central ideas and preoccupations of the emerging tradition of neo-institutional theory. Their key contribution was the decisive delineation between formal structures such as written rules, protocols, and laws officially adopted by institutional actors (specifically organizations), and the actual practice patterns used to perform effectively and efficiently on a daily basis (Powell & DiMaggio, 1991). They argued that organizations often adopt formal structures at the expense of effective and efficient functioning to signal commitment to prevalent and powerful socially sanctioned “rationalized myths” (Meyer & Rowan, 1977). So, for instance, by producing a collaborative vision statement and advertising it on their website as official policy, a healthcare organization signals that they are collaborative, even though attempting to realize this vision may be impractical.

Within this explanatory framework, actors, whether individual or organizational, align themselves with socially sanctioned ideas and norms by imitating the formal structures in which these ideas and norms are embedded (DiMaggio & Powell, 1983; Meyer & Rowan, 1977). The adoption, adaptation or mimicry of legitimized formal structures grants these actors legitimacy, as committed participants in realizing a shared, idealized view of their functioning. This may be particularly important during periods of uncertainty or resource constraint, when survival is, or appears to be, precarious (DiMaggio & Powell, 1983). By officially subjecting themselves to formal structures (or by being officially subjected to them) actors attain a ceremonially derived legitimacy, immediately communicating their ostensible worth (Meyer & Rowan, 1977).

This is all independent of how well formal structures mesh with or reflect the actual, practical work demands of the individuals or organizations who adopt them. The sphere of formal structure, specifying official attitudes and commitments is persistently “de-coupled” from actual patterns and needs of goal-directed functioning (Bromley & Powell, 2012; Meyer & Rowan, 1977). So, for example, structured initiatives like interprofessional rounds may continue to be developed and maintained for ceremonial purposes, despite failure to enable collaborative, patient centered care (Hill, 2003; Long, Forsyth, Iedema, & Carroll, 2006; Paradis, Leslie, & Gropper, 2016).

Analytic structure

The following analysis is organized in two major sections, each pertaining to one aspect of the collaborative healthcare ideal. In section one, we describe the “idealized collaborative clinician” and illustrate its resonance with a deep-seated modern ideal of a society composed of rational, autonomous and agentic individuals (Frank & Gabler, 2006; Frank & Meyer, 2002; Meyer et al., 1997; Meyer & Bromley, 2013). We then introduce and contextualize an element of formal structure that constructs one form of this “idealized collaborative clinician”, the CanMEDS Physician Competency Framework (Frank, Snell, & Sherbino, 2015) and examine how the language used in it achieves this construction. In section two, we describe the “idealized collaborative organization” and illustrate how this idealized construct echoes the deep-seated ideal of the modern organization (Meyer et al., 1997; Meyer & Bromley, 2013). We then introduce and contextualize two elements of formal structure, guides for collaboration in Family Health Teams (FHTs, pronounced “fits”) and show how their language constructs an image of multidisciplinary primary care in line with the ideal of the modern rational and autonomous organization (Meyer et al., 1997; Meyer & Bromley, 2013). The documents we examine are highly appropriate for an analysis of collaboration’s institutionalization through formal structure. Both played significant roles in major healthcare restructuring movements in Canada, in which collaborative care was a core priority: competency-based medical education and primary healthcare reform.

The idealized collaborative clinician

To begin, we describe what we refer to as the “idealized collaborative clinician.” We use this term to denote an ideal, or model, of what contemporary clinicians are or should be, and interpret it as an expression of a more general ideal central to modern culture: the autonomous, rational individual (Frank & Gabler, 2006; Frank & Meyer, 2002; Meyer et al., 1997; Meyer & Bromley, 2013).

Broadly speaking, feudal society was based on a starkly hierarchical social model. People’s identities, rights, obligations and capacities were more or less assumed to be fundamentally and unalterably unequal, a function of the social stratum they were born into (e.g. monarchy, aristocracy, clergy, peasantry). This stratified order was legitimized by religious myths, specifying the different strata, and the people within them, according to their proximity to the divine. Divinely sanctioned, this social construct was viewed as an expression of God’s will, and therefore timeless and optimal, immutable and best.

With the emergence of modern culture, this vision of society was supplanted by a new idealized image: a society composed of equal individuals possessing the same natural rights, freedoms and obligations, and the same basic capacity for reason.¹ Progressive improvement in knowledge and human life was to be made possible by allowing individuals to exercise their full natural rights, without infringement. Whereas the prior ideal organized itself around a divinely ordained and timeless hierarchical order, the new, modern society was envisioned as one composed of rational, autonomous, and agentic individuals (Frank & Gabler, 2006).

We argue that a key aspect of the collaborative healthcare ideal is that it overlays this idealized view of modern society onto the clinical domain – that is, the domain in which patients are provided direct care. We use the term “clinician” to denote a member of a healthcare profession or occupation who interacts directly with patients, providing them with healthcare services. These can include physicians, nurses, pharmacists, dieticians, occupational therapists, psychotherapists and others.

Within an idealized collaborative version of healthcare, the clinician (i.e. the abstract, general category) is analogous with the more general modern individual; the healthcare team is analogous to the progressive, just society. The idealized clinical team is composed of roughly equal, autonomous clinicians, each of whom provides care to the full extent of their roles and expertise and infringes minimally on each other’s roles and expertise, in the service of better patient care. Within this idealized depiction, the goals of coordinated activity are reliably achieved despite the absence or minimization of coercion. Instead, each and every actor (i.e. clinician) is, by definition, intrinsically committed to their role within the group and to the purposes for which the activity occurs.

In what follows, we examine how one specific form of the idealized, collaborative clinician, the collaborative physician, is constructed in the content of a particular element of formal structure: the CanMEDS Physician Competency Framework (Frank et al., 2015).

Institutionalizing the collaborative clinician: the canmeds physician competency framework

Collaborative ideas and values have in recent years been incorporated within a wide variety of formal structures attached to the education, training and certification of clinicians (Paradis & Whitehead, 2018). Chief among these have been competency frameworks, such as the Royal College of Physicians and Surgeons of Canada (RCPSC) CanMEDS Physician Competency Framework (Frank, 2005; Frank et al., 2015), and the Accreditation Council for Graduate Medical Education (ACGME) Core Competencies, developed in the USA (see Holmboe, Edgar, & Hamstra, 2016). These frameworks emerged at a time of concern that medical education was outdated, unsystematic, inconsistent, and insufficiently attuned to societal needs (Batalden, Leach, Swing, Dreyfus, & Dreyfus, 2002; Frank, 2004). To address these concerns, organizations responsible for setting physician residency standards, like the RCPSC and ACGME, developed competency frameworks to serve as generic foundations for training and certification of specialist physicians. Accordingly, each is structured around a small set of general “core competencies” (ACGME) or “roles” (CanMEDS), broad categories of skills, knowledge and attitudes (Frank, 2005; Frank et al., 2015; Whitehead, Austin, & Hodges, 2011). Since its development in the 1990s, CanMEDS has become one of the most influential competency frameworks in Canada, if not the world (Frank, 2005; Frank et al., 2015).

Produced by the RCPSC, the regulatory body responsible for certifying specialist physicians in Canada, it now “forms the basis for all Royal College educational standards for specialty education” (Frank et al., 2015, p. 5). The CanMEDS is now incorporated within residency program accreditation, evaluation and examinations, in addition to the standards for specialty training and continuing professional development (The Royal College of Physicians and Surgeons of Canada, 2019). Beyond specialist medicine, the College of Family Physicians of Canada (CFPC) developed a version adapted for family medicine (Shaw, Oandasan, & Fowler, 2017), which they subsequently incorporated into their own training standards and requirements. CanMEDS has also had a direct influence on formal structures developed and adopted by other clinical professions, and in numerous settings around the world (Frank et al., 2015; Whitehead, 2011). As such, CanMEDS has become one of the most prominent and influential formal structures in health professions education.

CanMEDS is organized around seven “roles”: medical expert, communicator, collaborator, leader, health advocate, scholar and professional (Frank et al., 2015; Whitehead et al., 2011). The content of the Collaborator role section (Frank et al., 2015, pp. 18– 19) depicts a physician who embodies the idealized model of the clinical team as a group of clinicians who are more equal than in the traditional model described, for instance, by Freidson (1970).

This idealized image is conveyed through depiction of the physician as one who works within the collective, in a manner that omits or minimizes coercion or command. In Key Competency 1 and its enabling competencies, the physician does not instruct but “works effectively with,” “establish[es] and maintain[s] positive relationships with,” “negotiate[s] overlapping and shared responsibilities with,” and “engage[s] in respectful shared decision-making with” clinical colleagues. In Key Competency 2 and its enabling competencies, the physician “work[s] with” colleagues “to promote understanding, manage differences, and resolve conflicts”, and “show[s] respect toward” them. While the physician’s position as grammatical subject (i.e. active agent) may still imply medical dominance (Paradis, Pipher, Cartmill, Rangel, & Whitehead, 2017), these formulations reflect a flatter interprofessional structure than the traditional one described by Freidson (1970).

The idealized collaborative physician is also constructed as an agent responsible for embedding collaborative norms within routine processes and structures, but fittingly, never in an explicitly coercive manner. Instead, the idealized physician acts to foster or “support” the self-evidently valuable objective of a “collaborative culture” through impersonal, rational means, by “[i]mplementing,” for example, “strategies to promote understanding, manage differences and resolve conflicts in a manner that supports collaborative culture.” Hence, the physician functions as an agent of collaboration in two senses: first, by acting to make the group more collaborative and second, by doing so in a non-coercive manner. The physician thus realizes collaborative precepts, not just by fostering collaborative group practices, but by embodying the collaborative injunction to avoid explicit coercion. In other words, this physician both engenders collaboration in external structures and processes and is collaborative.

This idealized construct of the physician and the physician’s work implies a corresponding idealized construct of the work group as a relatively flattened hierarchy. If the physician works with other clinicians then, according to the ideal, other clinicians are not given orders to follow, but are encouraged to work together with the physician. In the implied arrangement, all parties contribute to the collective effort as autonomous, empowered agents, whose freedom is bounded first and foremost by respect for the corresponding freedom and contributions of the other party or parties within a “collaborative culture”. Thus, by constructing the idealized physician role in this way, the collaborator competency domain also constructs an idealized version of the group and of every clinician working within it. This amounts to a putative demotion of the physician, and promotion of others, to a more level playing field, more in line with the corresponding modern ideal of a society composed of autonomous but rationally bounded individuals.

The idealized collaborative healthcare organization

Meyer and Bromley (2013) argue that, following World War II, in the increasingly globalized world, the Nation State’s idealized role as the most legitimate collective actor started to diminish, along with the legitimacy of the centralized State bureaucracy. Concurrent with this change, a new idealized form of coordinated social action emerged to fill the vacuum: the modern organization. This idealized form, they argue, has spread to an ever-increasing range of domains of human concern, becoming ubiquitous as one of the most basic assumed manifestations of coordinated group activity.

As an ideal, the modern organization consists of two key characteristics. First, analogous to the modern democratic state, it is an association of rational, equal individuals, autonomous but bounded by norms of mutual respect and commitment to the general social good. As such, it is internally empowering and minimizes internal coercion. Its coordination and effectiveness are functions of its members’ commitment to norms of mutual respect and shared social purpose. Second, the idealized modern organization, as a whole, is equivalent to the modern individual, an outward facing actor in its own right. As such, it is autonomous and distinct, chooses and rationally pursues its own mandates and self-interest, but is constrained by its respect for other actors and by its commitment to the pursuit of the general social good (see Meyer et al., 1997; Meyer & Bromley, 2013).

This idealized form of collective activity takes a particular form in healthcare: the collaborative healthcare organization. First, as noted earlier, the idealized clinical team within the organization is composed of independent, autonomous clinicians, with their own expertise, disciplines, understandings and interests. Rather than dissolving individual autonomy and distinctness, the idealized collaborative clinical team nurtures mutual understanding and respect for this autonomy and distinctness. The team is defined, moreover, by norms of common understanding and commitment to putatively self-evident social goods like better care and better patient health. By definition, it subordinates individuals’ self-interest to these social goods.

Second, the idealized collaborative healthcare organization itself is an autonomous but bounded actor which acts within a community of other similarly constituted organizations. Organizations within such a community are autonomous and distinct, in that they possess and pursue their own motivations, mandates and interests. The idealized inter-organizational partnership is a state of, and process of developing, norms based on mutual understanding and awareness of distinct interests and motivations. Each organization engages in introspection, individually and together with partners, allowing for the emergence of clear roles and values. This in turn allows for identification of, and commitment to, intrinsically worthwhile social goods that are visible in strategic plans, such as improving the lot of marginalized communities, or becoming a high-reliability organization.

In what follows, we examine how the idealized collaborative healthcare organization is constructed in two documents produced to support the creation and development of FHTs in Ontario (Ontario Ministry of Health and Long-Term Care, 2005; Quality Improvement and Innovation Partnership, 2009).

Institutionalizing the collaborative healthcare organization: the family health team model

During the 1990s and early 2000s, primary healthcare reform emerged as a major official priority of Canada’s federal and provincial governments (Watson & Wong, 2005).² This occurred in part as the fruition of a decades-long reform movement, and in part because of concerns for the sustainability and quality of the country’s single-payer healthcare system (Ritchie, 1997; The Conference Board of Canada, 2014; Watson & Wong, 2005). Within this context, in 2005 the Ontario government launched a program to implement a primary care reform initiative based on a multidisciplinary care model called the Family Health Team (FHT) (Ashcroft, 2015).

In the FHT model, care is provided by physicians and one or more other clinicians, such as pharmacists, nurses, dieticians, occupational therapists, psychotherapists or others working together to plan and execute patient care (Ashcroft, 2015; Meuser, Bean, Goldman, & Reeves, 2006; The Conference Board of Canada, 2014). The specific size and make-up of each team was intended to be flexible, depending on available human resources and the needs of the populations served (Ontario Ministry of Health and Long-Term Care, 2004). To incentivize and support the creation of, or transition to, FHTs, the government used a blended capitated physician remuneration scheme, payments for expanded scopes of practice, bonuses for achieving prevention targets, and committed to paying the salaries of all non-physician clinical team members (Rosser, Colwill, Kasperski, & Wilson, 2011).

The government also provided assistance and informational resources to aid the FHTs’ development and operations. Among the informational resources were the Guide to Collaborative Team Practice (Ontario Ministry of Health and Long-Term Care, 2005), and the Community Partnerships Resource Guide (Quality Improvement and Innovation Partnership, 2009). These two documents construct the FHT, respectively, as internally and externally collaborative.

The internally collaborative FHT

The Guide to Collaborative Team Practice (Ontario Ministry of Health and Long-Term Care, 2005) constructs the idealized FHT as internally collaborative: FHT clinicians collaborate across specialties within their team, making “mature” clinical teams “effective” and “well-functioning.” Prior to joining a FHT, the individual clinician is constructed as radically autonomous: she has and pursues distinct self-interests, purposes and understandings. This construct is salient in the document’s description of the “earliest stages” of team formation. Motivated by self-interest, and “dependent on identified leaders,” individuals compete to “outperform each other or to gain attention from leaders or supervisors” (p.8). At this stage, they still envision themselves as embodying independent personal and professional identities, defined through education. As a consequence, when the team starts to form, they “may be unsure as to their roles [and] the purpose of the team” (p. 8).

By contrast, in the idealized, “effective”, “well-functioning” or “mature” FHT, individual clinicians’ autonomy is preserved, but within defined bounds, specifically “negotiate[d] roles”. The “[r]oles and responsibilities of all team members need to be clear, explicit and understood by all other team members” (p. 6). Accordingly, team members “learn about each other, their roles and the contributions they can make” (p. 6). Moreover, the team is governed by norms of “respect for each-other’s professional and personal strengths and limitations, valuing of diversity and confidence in each other’s ability to achieve the team’s goals” (p. 7). Hence, within this idealized team, individual members’ personal autonomy is defined within clearly articulated, distinct roles, negotiated within the team and, in turn, realized through norms of mutual respect.

The document also circumscribes clinicians’ autonomy within the bounds of team goals, which are rooted in putatively self-evident social goods. Explicit coercion, through extrinsic forces like financial incentive, punishment or censure are minimal. Like team members’ individual roles, the teams’ collective purpose and goals are to be made rational and explicit, such that they are “clear and understood by all” (p. 6) so that they can be internalized. These goals, rooted in legitimized purposes such as improved patient health, are to supersede all selfish interests. The mature team “focus[es] on meeting the needs of the population(s) being served rather than on issues related to individual team members” (p.6). While individual members work from within distinct and autonomous roles, as a team approaches the ideal, their “individual contributions increasingly [are] channelled towards [these] common goals” (p. 8). There is thus an implicit, progressive subordination of personal needs or aims to shared, group goals.

The externally-collaborative FHT

The Community Partnerships Resource Guide (Quality Improvement and Innovation Partnership, 2009) constructs the FHT as externally collaborative: the idealized FHT collaborates with other health and community organizations and constitutes an actor in the construction of an integrated, inter-organizational healthcare system.

The Community Partnerships guide was produced with the expressed purpose of helping FHTs to form and maintain partnerships with other organizations. In doing so, it transforms the mythologized FHT into a radically autonomous “self” in its own right, in a rhetorical move that is analogous to the construction of the idealized collaborative clinician. The document explicitly invokes Tuckman and Jensen’s (Tuckman & Jensen, 1977) stages of small-group development model, which characterizes the early stage of development as “chaotic”, with different organizations “vying for leadership” and “not checking out assumptions” (Quality Improvement and Innovation Partnership, 2009, p. 14). The document stresses also that each individual organization involved in a partnership, including a FHT, pursues its own individual advantage, survival and “welfare” extending even to its involvement in the partnership, such that it “may even have joined the partnership to protect against a perceived threat” (p. 14).

In the idealized partnership depicted in this document, the autonomy of the FHT is bounded by that of its partners and vice versa. The negotiation of boundaries occurs continuously with increased understanding and awareness of difference between the FHT itself and its partners. Within the FHT itself, awareness of its own identity and particularities is to be achieved through self-reflection, followed by externally directed communication of what is discovered. With partners, mutual awareness comes from “taking time to find out more about each partner’s interests” (p. 15). The language reflects the rational actorhood of each member of the idealized partnership, since each “examine[s] [its] own thinking- [its] pathways of assumptions and observations that support [its] beliefs...” and “learn[s] how to make these pathways more visible to each other” (p. 15). On this basis, the inter-organizational partnership is analogous to that among team members, who can achieve “clarification of expectations [and] attention to roles and responsibilities” (p. 14), and therefore establish a foundation for autonomous and bounded contributions.

Another important characteristic of the externally collaborative FHT is its commitment to socially legitimized purposes, such as patient and community well-being. These commitments constrain selfish motivations by “[t]aking account of the community’s well‐being as well as [its own] self‐interest” (p. 14). In preparing for the partnership, the idealized FHT reflects on fundamental “principles and values” of universal importance, such as “[c]ommitment to marginalized and low‐income populations,” “[r]esponsiveness to community needs,” and “[r]eflecting diverse communities” (p. 16). By “keep[ing] a focus on what the partnership is about” (p. 15), that is, its core values, divergent purposes are constrained by commitment to deeply legitimized values of social utility such as equality, community and diversity.

Discussion

In this article, we have illustrated the construction of a form of the idealized, collaborative clinician (the collaborative physician) as well as the construction of the idealized collaborative healthcare organization, within two elements of formal structure: the CanMEDS Physician Competency Framework and two guides for collaborative practice in FHTs. We have argued that these examples of formal structure illustrate an idealized representation of a society free of coercion, composed of autonomous but bounded, ethical and purposive modern actors (Meyer & Bromley, 2013).

As elements of structure, these documents are intended to constrain or direct human activity in the particular practical domains in which they are introduced: CanMEDS is incorporated within education, testing and certification; and the FHT guides are incorporated within application and development processes for primary care teams. Unlike position pieces or abstract models of care, these documents are organized and introduced as intended mechanisms within larger causal sequences bringing about better clinical care and better health system organization. They are produced for practical use and their ideas about collaboration are integrated within the structures and processes of clinical training and health system restructuring. As such, they do not merely express the idealized view of collaboration but also embed and construct it. They make the idealized view of collaboration “real” in people’s minds and in organizational forms, but – and this is key – not necessarily in practice. They make room for what neo-institutional scholars have called policy-practice decoupling: the gap between formal policies and the practices that may or may not map onto them (Bromley & Powell, 2012; Meyer & Rowan, 1977; Paradis, 2017).

There are two key implications of the embedding of the collaborative ideal in formal structures. First, these idealized constructs create unavoidable pressures and constraints on actors within the system as routine aspects of their everyday lives (Berger & Luckmann, 1966). To be sure, this effect may extend beyond individuals, settings and organizations formally specified as “collaborative” (e.g. FHTs). When embedded within larger coercive structures like certification, accreditation or funding, idealized constructs stand to exert powerful pressures for individual and organizational actors, wherever these regulatory mechanisms reach. The need to demonstrate actual or superficial conformity with these constructs may be a prerequisite for a whole range of individual and organizational actors to achieve and maintain their positions within the system (Meyer & Rowan, 1977). These idealized constructs also exert powerful influence through less explicitly coercive structures, like predominant norms and available models of professional roles and organizational form (DiMaggio & Powell, 1983; Meyer & Rowan, 1977).

The second implication of the embedding of collaborative elements within existing healthcare structures is that this creates the impression that they already co-constitute the basic furniture of institutional life, and that they are not merely goals to achieve. They are thus constructed as real in the sense that they seem to represent what is actually done and can be done readily, in practice. The constructs come to be internalized or taken for granted as stable elements of mundane reality (Berger & Luckmann, 1966). For instance, the idealized collaborative clinician is incorporated, via CanMEDS, into mandatory certification criteria. Because all actual physicians (i.e. those allowed to practice) have been certified, we believe that, or act as though we believe, that they can and do collaborate in practice. If a group of practitioners goes through the formal application and development phase and is “awarded” FHT status, we come to think, speak, or act as though they are a self-enclosed, internally-and externally-collaborative “team” and are, in practice, aligned with the corresponding model and ideals of a team. Yet in practice, they may be aligned only superficially with team ideals, for example if they share a building but rarely interact. Or they may be collaborative in some of their practices but not others, working toward greater collaboration or moving toward less.

Hence, the presence of deep-seated societal beliefs within formal structures perpetuates what Meyer and Rowan (1977) refer to as a “logic of confidence”. It not only perpetuates an idealized account that collaborative care is best but embeds this account in structures that, by appearance or convention, seem to represent how the system works. The “blueprints” appear to be more than just tentative representations of what should be achieved, but instead appear as factual representations of the system and its elements. Within this objectivated (Berger & Luckmann, 1966) representation, clinicians and organizations, as basic elements of the healthcare system, appear to embody the idealized account that frames them as collaborative. This is independent of whether or not the formal structures in question provide sufficient constraint to yield actual conformity in everyday activities. For the logic of confidence, it is sufficient that the ideal has been laid out in formal, articulated, seemingly practical structures and processes; no further proof is needed than mere existence, since we trust in these formal structures to accurately represent reality.

We argue that this interpretation helps to explain the disconnect between the exuberant enthusiasm associated with the ideal of a collaborative healthcare system, and the fact that efforts to realize it have yielded such apparently modest results on patient outcomes (Reeves et al., 2017, 2013). Where others have pointed to faulty structures or power hierarchies, we have emphasized the normative underbelly of collaboration, and noted how collaborative formal structures resonate powerfully with deep-seated modern sensibilities. The language of the collaborative clinician and the collaborative organization reflects a shared narrative about the intrinsic superiority of democratic, minimally coercive social structures, and the empowered individuals conceived of as their basic elements. Embedding collaboration within the formal structures of healthcare reinforces this narrative in a way that mere rhetoric never could. It convinces us that these modern, democratic ideals are already integral to the system and how it works. It convinces us that these ideals are defining, essential, indispensable features of the routine, stable, taken-for-granted reality of healthcare. Hence, collaborative interventions are tried and tried again, even when empirical justification is lacking, because the justification is taken as self-evident.

Conclusion

In this article we have presented a new perspective on the discourse and practice of collaboration, which is admittedly empirically limited. Future empirical work inspired by institutional theory or using institutional ethnography is warranted to elucidate the gap between collaborative formal structures and their enactment in practice. Subjects of future research should include the identification of barriers and facilitators to realizing collaborative ideals in practice. One key topic is how well collaborative ideals and formal structures containing them account for the range of constraints on clinicians’ and organizations’ routine work. Other salient topics include how collaborative formal structures are implemented in practice at all organizational levels, how incorporation is demonstrated and how these processes can be improved. Also warranted is a comprehensive account of collaboration’s institutionalization within healthcare through a wide range of formal structures, using the general framework employed here. Empirically grounded analyses of this kind can help to explain the rise of the collaborative ideal to widespread acceptance and its influence on healthcare reform efforts.

Declaration of interest statement

The authors report no conflicts of interest.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Acknowledgments

The authors wish to thank Aria Birze, Madison Brydges, Andrea Carson, Patricia Leake and Victoria Whyte for comments on previous drafts of this article. The content is the sole responsibility of the authors.

Additional information

Funding

This work was supported by the Social Sciences and Humanities Research Council (SSHRC) through an Insight Development Grant (# 430-2016-00927) and the Canada Research Chair in Collaborative Healthcare Practice.

Notes on contributors

Daniel W. Miller

Daniel W. Miller is an independent Research Associate, working at the Leslie Dan Faculty of Pharmacy, University of Toronto, Toronto, ON.

Elise Paradis

Elise Paradis is Assistant Professor (status only) at the Leslie Dan Faculty of Pharmacy, University of Toronto, Toronto, ON.

Notes

1. In truth, these ideals, and the early social reforms in which they were encoded (e.g. property rights, voting), still excluded large segments of the population.

2. Canada has a single-payer health insurance system, in which the federal government distributes funding to the provinces who then administer their own insurance programs. Services are not provided directly by public employees but by providers operating independently or within private organizations.