Abstract
Dougherty et al. posit that production of complex innovations requires that ecologies be organized, involving three activities: orchestrating knowledge capabilities, ongoing strategizing to frame and direct continuous innovation, and developing public policy to embrace ambiguity. Our study aims to understand how such ecologies emerge. Based on a longitudinal case study, performed in the context of the Quebec health system, our results suggest (a) that the emergence of innovations in highly institutionalized fields requires an additional activity, namely, working on boundaries to make actors perceive their interdependences (b) some levers that can foster the implementation of the model.
Keywords
Introduction
Under pressure by demographic, epidemiological, and technological factors, all Western countries strive to control costs and improve the quality and comprehensiveness of health care services. Innovation is considered as critical for the sustainability of health care systems (Casebeer et al., 2006). Innovation is defined as a process that involves the generation, practical adoption, and spread of new and creative ideas intended to produce a qualitative change in a specific context (Sorensen & Torfing, 2011). Because of the complexity of health care needs, innovation processes involves several interdependent organizations and professionals; the aim is to provide more integrated solutions to often interconnected problems (Albury, 2011). Consequently, innovations in governance are at the heart of the transformation of the systems (Denis, 2002). As stated by Moore and Hartley (2008), innovations in governance may logically entail or create conditions under which many different service and production processes innovations can occur. Innovations in governance differ from innovations in products, services, and processes in at least two important senses. First, the innovations are conceived and implemented above the organizational level: They involve networks of organizations, or the transformation of complex social production systems. Second, these innovations focus on concrete changes as well as on the ways in which productive activity is resourced, the processes used to decide, and the normative standards used to evaluate the performance of the social production system. Briefly stated, these innovations in governance are related to changes in the way interdependent actors coordinate their collective action (Hatchuel, 2000). Moore and Hartley argue that innovations in governance have been an important part of overall innovation in government, and thus should be analyzed as much as service innovations.
This being said, the production of such innovations is particularly difficult due to the inertia and the fragmentation of the system (Coiera, 2011). To overcome these difficulties, two complementary strategies have been proposed. The first consists in designing spaces for innovation intended to protect actors from “institutional discipline,” enhance creativity, and experiment with new solutions (Zietsma & Lawrence, 2010). The second strategy aims to exploit the potential of multiactor collaboration to spur public innovation. Collaboration between relevant actors can strengthen the constitutive cycles of innovation (Sorensen & Torfing, 2011): generation of ideas (mutual learning), selection of ideas (compromise and agreement in decision making), implementation of selected ideas (mobilization of resources), and dissemination of innovative practices (formation of social and professional networks).
Despite the potential of these two strategies, their real impact in terms of practices’ change is questionable. Empirical studies show that networks do not always foster collaboration, and collaboration does not always lead to innovation (Sorensen & Torfing, 2011). While spaces for innovation can produce innovative ideas and concepts, implementation challenges remain (Cartenson & Bason, 2012; Grenier & Denis, 2011). Consequently, there is a need for contextual investigation into how spaces for innovation based on collaborative processes transform practices.
Dougherty and Dunne (2011) posit that production of complex innovations, involving multiple interdependent actors, requires that ecologies be organized. Specifically, they propose that
innovations can continually emerge productively if people work locally in ecologies to set and solve problems of orchestrating knowledge capabilities across the ecology, strategizing across the ecology to create new businesses and applications, and developing public policies to embrace ambiguity. (Dougherty & Dunne, 2011, p. 1214)
This model is interesting because it emphasizes organizational issues (such as development of capabilities) akin to change practices, but as mentioned by the authors, we need to understand empirically how these ecologies emerge. Our ambition is to contribute to this understanding and especially to validate the relevance of the model proposed by Dougherty and Dunne for the understanding of the emergence of innovations in highly institutionalized sectors. This article aims to answer a central question: How does a space for innovation, that is, using what levers (structures, resources, incentives, practices, etc., allow an ecology to emerge in a sector characterized by a high-level inertia, namely, the health care sector? More precisely, how can a space for innovation foster the orchestration of knowledge capabilities, strategizing, and the embracement of ambiguity?
We examine the implementation of an innovation space in the Quebec health care sector, called a Knowledge Sharing and Skills Development Initiative (in French, Initiative sur le partage des connaissances et le développement de competences or IPCDC). The IPCDC aims to introduce degrees of freedom in the system to allow innovation in governance, and to support the implementation of comprehensive intervention approach for health problems, namely, the population-based approach.
This article is structured as follows. The first part (Background) presents the main streams of thought concerning the potential of collaboration-based spaces for complex innovations and our theoretical framework; the seconds section (Method) outlines our methodological approach. Section Three (Results) presents the results of the empirical study of IPCDC implementation. The conclusion (Discussion and conclusion) discusses issues surrounding the emergence of ecologies of complex innovation.
Background
Three research streams provide interesting insights into the potential of spaces for complex innovations based on collaborative processes: (a) works relying on neo-institutional theories that try to explain innovations in highly institutionalized environments, (b) works related to collaborative innovation, and (c) works concerning complex innovation processes. The first two streams clarify how the creation of a space and networking can produce innovation. The third stream focuses on the specific challenges related to the production of complex innovations and proposes a model for organizing ecologies.
How Can Embedded Agency Produce Institutional Change?
Some authors argue that institutional arrangements form a kind of iron cage in strongly institutionalized settings (DiMaggio & Powell, 1983) that impede innovation. Institutions are social structures (cultural-cognitive, normative, and regulative elements) that, together with associated activities and resources, impart stability and meaning to social life through pressure to conform (Scott, 2008, p. 48). In recent years, studies on institutional change have challenged this traditional thinking (Reay & Hinings, 2009), seeking to reconcile exogenous and endogenous (Boxenbaum, 2004) explanations for institutional change. The exogenous explanation suggests that change occurs when a precipitating jolt, in the form of an external event, disrupts, and alters the existing institutional order. The endogenous explanation suggests, conversely, that change arises from human interaction. Adopting this perspective obliges one to address a paradox of embedded agency: “How can actors change institutions if their actions, intentions, and rationality are all conditioned by the very institution they wish to change?” (Holm, 1995, p. 398). Scholars have advanced several responses, corresponding to three approaches:
The first approach emphasizes the role of actors (called institutional entrepreneurs) who “see in new institutions an opportunity to realize interests that they value highly” (DiMaggio, 1988, p. 14), and exploiting precipitating jolts to challenge existing institutional arrangements. Several studies have shown that these entrepreneurs are often in peripheral positions or are new entrants or members whose positions span multiple fields. Such positions enable agency. Furthermore, numerous works (see, for example, Wijen & Ansari, 2006) suggest that many institutional changes require collective institutional entrepreneurship, involving “collaboration” among dispersed actors.
The second approach construes institutional change as the outcome of the dynamic interactions between two institutional by-products (Seo & Creed, 2002): institutional contradictions and human praxis. Contradictions can arise when products of institutionalization processes, which often reflect the ideas and goals of the most powerful actors, are unlikely to satisfy the divergent interests of all participants. The ongoing experience of these contradictions transforms actors from passive participants in the reproduction of existing social conditions to mobilized change agents. Seo and Creed (2002) assert that human praxis is the core mechanism of institutional change, through three components: (a) a critical understanding of the existing social conditions, (b) actor mobilization, and (c) collective action to reconstruct existing social arrangements. An empirical study by Zietsma and Lawrence (2010) shows how the transition from conflict to innovation was triggered by a combination of practice work and boundary work. Boundary work and practice work refer to actor’s efforts to, respectively: (a) establish, expand, reinforce, or undermine boundaries; (b) create, maintain, or disrupt practices. Institutional innovation appears to be enabled when boundaries around experimental spaces protect projective agency from institutional discipline (p. 214).
The third practice-based approach examines how everyday activity can be the locus for changes to practices that are anchored in field-level institutional logics. Lounsbury and Crumley (2007) explain how variation in activities can spur field-wide efforts to establish an innovation as a practice. Using the notion of performativity (Feldman, 2003), they advance an interesting interpretation of institutional change as an endogenous process. Performativity emphasizes that “activity is often accomplished by skilled actors who rely on practical evaluative agency to understand and assess how practices can be altered or tailored in order to accomplish specific tasks or to cater to different audiences” (p. 999). Their case study interprets the innovation process as a distributive process that involves theorization and legitimation of a new practice (Lounsbury & Crumley, 2007).
The above works suggest that innovation processes are characterized by the distributed agency of actors permitting collective action for innovation. This agency is anchored in their daily practice and depends on ongoing contradictions at the field level. Specific contexts such as crises at the field level or pressure at the activity level and different means such as experimenting (practice work), theorizing, actors’ mobilization, and constructing new boundaries (boundary work) may help undermine inertia. We will now discuss the collaborative context of such collective action.
Do Collaborative Processes Produce Innovation, and if so, How?
Collaborative processes can be levers for innovation. As underlined by Hartley, Sorensen, and Torfing. (2013, p. 825), empirical evidence is supported by arguments about how collaboration can strengthen all stages of innovation: The framing of problems is often improved when different experiences are brought together; the generation of creative solutions is enhanced when different ideas are developed, combined, challenged, and built on; the selection and testing of promising ideas is strengthened when diverse actors help assess gains and risks; the implementation of new solutions can be improved when different resources are mobilized, exchanged, and coordinated; and finally, innovative ideas are diffused when collaborators become external ambassadors for new ideas and practices. However, several factors can raise obstacles to collaboration, such as cognitive and identity barriers (Ferlie, Fitzgerald, Wood, & Hawkins, 2005), incongruent goals, power asymmetries (Gray, 1989), ideological conflicts, and mistrust (Gray, 1989). In addition, even when actors collaborate, innovation may be hampered in contexts where there are coordination problems (Swan & Scarbrough, 2005), where networks are too stable and closed (Skilton & Dooley, 2010), where there is a high level of strategic uncertainty (O’Toole, 1997), and where there is risk aversion (Hartley et al., 2013).
Conversely, Davis and Eisenhardt (2011) suggest that three mechanisms tend to characterize successful collaborative innovations: (a) the activation of relevant capacities, (b) a deep and broad innovation search trajectory, and (c) mobilization of diverse participants over time, which requires a rotating leadership. However, a rotating leadership does not always lead to innovations in complex contexts, characterized by a strong interdependency between actors. Gherardi and Nicolini (2005), looking at accounting practice innovation in the Italian local administration, propose that these complex processes are better understood through an ecological perspective that emphasizes the context, the reciprocal change of actors and situations, and the negotiated nature of social order (even when negotiations are not apparent). The next section discusses this ecological perspective in detail.
How Can We Deal With the Specific Issues of Complex Innovations?
Producing complex innovations, involving different interdependent organizations, is challenging because it requires organizations to develop and maintain collaboration for long periods of time with a high degree of ambiguity. Dougherty and Dunne (2011) posit that organizing ecologies of such complex innovations is needed to effectively enable and shape emergence of innovation. These authors elaborate the dynamics of emergence (Dougherty & Dunne, 2011, p. 3) from an extensive review of works inspired by complexity science. First, emergence can happen only if there are enough connections between agents—people, organizations, and knowledge systems—to allow new patterns to emerge. Second, deviation-amplifying activities such as positive feedback move the system toward a new kind of order. The third dynamic enables new orders to come into being and comprises coordinating mechanisms that recreate, reuse, and recombine existing elements into a new order that increases the system capacity. Based on this understanding, Dougherty and Dunne propose a model specifying the activities that must be organized in the ecology for innovation. They emphasize the importance of three main activities: orchestrating knowledge capabilities, ongoing strategizing, and developing public policies that embrace ambiguity.
Orchestrating knowledge capabilities to support innovation
Fostering the three dynamics of emergence rests on (a) linking knowledge capabilities with one another and with “market” needs so that new insights can emerge, (b) transforming knowledge at boundaries, and (c) focusing on actual product innovation activities, which become a common ground (Dougherty & Dunne, 2011). Mechanisms such as standard-setting bodies and R&D consortia may support such knowledge orchestration.
Ongoing strategizing to frame and direct continuous product innovation
Implementing the three dynamics of emergence involves specific modes of organization: (a) bundling capabilities into businesses and matching these businesses with market opportunities, (b) negotiating strategic direction in real time by focusing on processes, and (c) making a long-term commitment (Dougherty & Dunne, 2011). Mechanisms such as strategic alliances extended to a large variety of constituents and specific governance arrangements to serve the interests of all stakeholders may support such ongoing strategizing.
Developing public policy to embrace ambiguity
Public policy plays an important role in innovation management. Specifically, development of public policies enables alliances between organizations and increases value appropriation (Dougherty & Dunne, 2011). Fostering the three dynamics of emergence entails (a) sustaining re-architected relations (e.g., between industry and research), (b) generating the deviating amplifying actions (e.g., institutional entrepreneurship) that allow new kinds of ideas and institutions to emerge, and (c) stabilizing new rules of the game (which may evolve as the game changes). Public agencies and other bodies, such as professional associations, are key players in supporting new public policies.
This model is preliminary. As Dougherty and Dunne (2011) note, further empirical research is needed to answer various questions: How do such ecologies emerge? And what kinds of practices and structures can foster collective learning among a large number of interdependent actors “working” in very different organizations and contexts (Dougherty & Dunne, 2011)?
We mobilize the promising work of Dougherty and Dunne (2011) to analyze how a space for innovation can favor the emergence of ecologies of innovation. We elaborate the following general guiding research question:
But given the specificity of the health care sector, a field highly institutionalized and fragmented that renders innovation much challenging, and inspired by other works (Beck & Plowman, 2014), we have chosen to embrace the understanding of the phenomenon of innovation with all its complexity, and as a consequence, we will not make specific propositions concerning the levers of change. The results of our study will be discussed further in light of other findings.
Method
Research Strategy
Trottier and Denis (2012, 2014) performed a longitudinal case study of the implementation of IPCDC. Case study is a very appropriate research design when the researcher asks why or how research questions and is interested in examining contemporary phenomena in their naturally occurring contexts (Stake, 1995). More precisely, IPCDC constitutes an instrumental case study. As explained by Stake (1995, p. 4), regarding selection of cases, “case study research is not a sampling research.” Our first obligation is to understand this one case. Therefore, the first selection criterion should be to maximize what we can learn (Stake, 1995, p. 4). Because of its context (health care system) and its object (creation of a space for innovation), this single case is of very special interest for the understanding of the emergence of ecologies of innovation in highly institutionalized milieus, which make change challenging.
Data Sources
This study is based on three main sources of data. First, interviews (45), performed essentially during 2011-2013, were conducted with (a) 30 representatives of the partner network (presented above) involved in the implementation of IPCDC, their collaborators (e.g., regional agencies), and the members of the operational team responsible for execution of the activities; (b) 15 representatives of health organizations that have participated in the activities organized by the IPCDC. The second source includes documents, such as activity reports from the IPCDC and minutes of strategic and executive committee. The third source consists in nonparticipant observations of meetings of different committees.
Data Analysis
We reanalyzed the data gathered in this implementation study to grasp how IPCDC has favored the emergence of an ecology of innovations in governance. More precisely, we examined the impact of IPCDC at the system level, and at the local level, which correspond, respectively, to changes in the way regulatory organizations (regional agencies, health department, etc.) and the health organizations and their local partners develop new governance arrangements to support the implementation of the reform.
These data were subjected to thematic analysis (Braun & Clarke, 2006), which consists in identifying and analyzing themes revealed by the data. We used an inductive and deductive approach that allowed us to derive meaning using our analytical framework while seeking new meaning. This method comprises five steps: familiarization with the data, coding, research-revision, definition of themes, and final report. Concretely, we raised the following questions to embrace activities proposed by Dougherty and Dunne (2011):
About orchestration of knowledge capabilities: What the different actors learned, individually and collectively, concerning more precisely the new approach (population-based approach) promoted by the reform? How these collective learning developed?
About ongoing strategizing: Here, strategy is conceived as a pattern, in a stream of actions (Mintzberg & Waters, 1985). This definition focuses on the behaviors of the actors. Some questions help us to grasp ongoing strategizing: How the organizational strategy of different actors evolved through time? How the innovations in governance affected the organizational strategy of actors? How the health organizations enacted the population-based approach?
About embracing ambiguity: Here, we are more interested in the way innovation risks are managed by actors. More concretely, the principal question is, how do actors deal with the fact that they ignore the outcomes of their actions?
The validity of the study
The means for enhancing internal validity or credibility (i.e., the plausibility of interpretations elaborated), and external validity or transferability (i.e., ability of the investigation to make sense elsewhere) of this study are specific and stem from the characteristics of our methodological approach (Mukamurera et al., 2006). To reinforce the credibility of this investigation, several strategies have been adopted such as emphasis on the methodological coherence between the research questions and the selected methods, triangulation of methods for generating data, concurrent processes of data collection and analysis, and so on. On the other hand, the transferability of the investigation is strengthened with the following strategies: an in-depth description of the case study in order to the reader be able to better appreciate the influence of the context; and our interest and effort in not only being strongly theoretically founded but also thinking theoretically throughout all the planned research steps.
Research Context
The case study draws on the Québec health care experience. The health care services are provided by public institutions offering primary, secondary, and tertiary health services, and by physicians in private or public practice, paid mostly on a fee-for-services basis. During the study period (2011-2014), the system was regulated on two levels:
The macroscopic level: The provincial government determined the extent of coverage under the medical insurance scheme, the population insured, the division of power across decision-making levels, and the definitions of the areas of professional practice.
The mesoscopic level: Regional Health Authorities (RHAs) were responsible for allocating resources to public institutions based on previously determined budgets. Health care organizations “signed” management agreements with RHAs, which are accountable to the Ministry.
In 2004, a major reform was introduced. The producer-oriented system, which aimed to return patients to a state of non-illness, respond to clientele demands, and tackle a determinant of health (notably health services), was replaced by a population-based approach, aimed at preserving health and its determinants, taking responsibility for a geographically determined population, carrying out a proactive strategy, and involving stakeholders.
This reform was implemented through the creation of 95 Health and Social Service Centers (HSSCs). These HSSCs resulted from the merger of several health care organizations operating within the geographical territory. The HSSCs were mandated to develop collaborative or contractual agreements with autonomous providers in their territory that offered services to the local population (e.g., voluntary agencies, medical clinics, and tertiary-care hospitals), as well as supraregional entities, to create local health and social services networks.
The implementation of a population-based approach faces numerous challenges (Breton, Pineault, & Lamarche, 2005): knowing the health needs of the population and establishing priorities for intervention; ensuring a continuum of care and services to strengthen interorganizational and interprofessional collaboration; reinforcing primary care, notably through better collaboration with private organizations; changing health service utilization habits; managing the interterritorial mobility of health users; acting upstream, that is, developing prevention services and targeting all the determinants of health, notably through improved collaboration with different public administration sectors (housing, education, etc.); finding new ways to reach vulnerable groups with complex needs who do not ask for services; and introducing new management practices, notably results-based management.
Consequently, the introduction of this reform implies an important change in governance: Improving population health is possible only if interdependent actors working in very different organizations and sectors collaborate at different levels (strategic and operational).
Results
In this section, we will describe IPCDC’s implementation, and analyze how it has contributed to the emergence of innovations in governance.
Implementation of the IPCDC: A Brief Description
This section describes the implementation of the IPCDC in three phases (Denis et al., 2012)
Mandated change as a trigger for innovation (2004)
The delegation of some public health responsibilities to local organizations was seen as an opportunity for public health organizations and actors to strengthen the institutional development of this sector, because it could create an interface between public health approaches and service delivery.
Although such changes were well received by public health leaders, they presented capacity and competency challenges for many HSSCs. Many of these organizations lack dedicated resources or expertise in public health.
Consequently, some key figures in the public sector (CEO of the National Institute of Public Health [NIPH]), a physician with a long career in public health, two NIPH directors involved in health systems and innovation issues, the director of a university’s community health department, and a former health minister who specialized in public health considered an initiative to help HSSCs assume their new responsibilities. Personal contacts helped secure the support and expertise required for the development of the initial design of the IPCDC.
Launching of the IPCDC and first cycle of activities (2005-2009)
In March 2005, the first meeting of partners and stakeholders was held to discuss the creation of a network to support the integration of public health responsibilities within HSSCs. They decided to create the IPCDC, a consortium composed of the Quebec Association of Healthcare Establishments; RHAs, including the public health department, the Ministry of Health and Social Services; the NIPH; the Quebec Observatory of Local Health Networks; and representatives of universities active in the field. Although the consortium incorporated a large set of partners, important segments of the health system, notably physicians’ unions and labor unions, were excluded during this phase. This transboundary consortium had no formal authority and acted in parallel with the structures of the system.
The initiative was officially launched in 2005 and has evolved steadily since then. Its setup consisted mainly in formalizing the steering structure, consolidating an activity program and a budget planning process, producing a communication plan (in 2007), and initiating an evaluation strategy (in 2008) to assess its viability.
During this phase, the steering of the IPCDC centered on the role of the strategic orientation committee, chaired by one of the leaders of the initiative and composed of representatives of the health care network and university partners.
The IPCDC’s initial activity program focused on four areas: (a) training local health agents in basic public health competencies and in change management; (b) organizing seminars (reflection groups), mostly with HSSC managers; (c) providing training sessions in community development; and (d) creating a “virtual” desktop to provide practitioners with resources in public health and health policy and management. In 2008, the IPCDC decided to prioritize competency development around population health responsibility.
The IPCDC is financially supported by a grant from the Ministry of Health and RHAs, renewable every 3 years with the support of other agencies (e.g., from the NIPH).
One of the key issues during this period was to secure and formalize support from the Quebec Ministry of Health and from RHAs, because the IPCDC was not formally part of the institutional landscape of the system. At the end of this period, the IPCDC faced some challenges: It did not have guaranteed funding beyond 2009, and it needed to demonstrate the progress of its activities and impacts.
Consolidation of the IPCDC (2009-2013)
From 2009, the IPCDC was urged by the Ministry of Health to formalize its steering structure to enhance its accountability. This culminated in the creation of a formal executive bureau for the IPCDC, which has become the seat of the IPCDC’s steering functions. The IPCDC’s strategic orientation committee was dissolved. The creation of the executive bureau served multiple purposes. It recruited members from key partners’ organizations that are involved with the realization of the IPCDC’s activity program, and helped develop synergy across the different initiatives and resources of the organizations that made up the consortium. The members of the executive bureau met regularly.
This period (2009-2013) was also marked by a major strategic shift in the program. The Ministry of Health encouraged the consortium to broaden its vision of the reform implementation issues. Consequently, the IPCDC, which had initially focused more on individual competency development through training, elaborated a more systemic strategy, and gradually adopted a more collective approach. In 2011, the Ministry formalized a strategic plan that stressed the importance of developing a coherent vision and supporting practice change in the health services system (Ministère de la Santé et des Services sociaux [MSSS], 2011) to strengthen the implementation of a population-based approach.
Starting from this period, a cohort of managers working in the same organization (or in organizations serving the same region) was offered training sessions, notably in change management. Several tools, such as a guide for change management and a framework of competencies, were also designed to support the implementation of a population-based approach. Finally, during this period, the pivotal activity of the IPCDC consisted in offering coaching programs to HSSCs that were implementing projects in line with their population health responsibility. Some of these projects were very ambitious because they involved several partners and sectors.
Concerning funding sources, the IPCDC received support from 2009 to 2012 from the Ministry of Health, RHAs and the NIPH. In 2012, the Ministry of Health curtailed guaranteed funding from 3 years to 1 year, partly due to budget constraints within the system. Also during this period, the IPCDC faced pressure from the Ministry of Health to “prove its added value.” The alliances between IPCDC and the Ministry eroded because of certain leaders’ departure and changes in the government.
Tables 1 and 2 present, respectively, the evolution of funding for IPCDC and a review of activities.
Funding Sources for IPCDC.
Note. IPCDC = Initiative sur le partage des connaissances et le développement de competences; NIPH = National Institute of Public Health.
IPCDC’s Activities.
Note. HSSC = Health and Social Service Centers.
The Emergence of an Ecology of Innovation: An Analysis
As we have shown, the IPCDC prompted significant changes in governance, at the system level. Responsibility for the reform implementation is shared between several interdependent actors, working in networks: the ministry and RHAs, including departments of public health; the Quebec Association of Healthcare Establishments; and representatives from universities active in the field. These actors have different visions about reform innovation issues. This approach is innovative because (a) it taps into a wide set of capacities and resources to support reform implementation, (b) it aims to influence decision making through negotiation and collective learning, and (c) it strives, notably through university representatives and evidence, to introduce new criteria for the evaluation of a strategy of reform implementation. To extend our analysis, we will examine how IPCDC allows an ecology to emerge, referring to the organizing principles proposed by Dougherty and Dunne (2011).
Orchestrating knowledge capabilities to support innovation
The analysis focuses here on collective learning issues.
Our data analysis suggests that coordination of knowledge development is supported by frequent meetings of the executive bureau and exchanges between participating members, thus fostering a common vision of the population-based approach:
When I came to the executive bureau, I didn’t see that population responsibility had ramifications in all clinical departments. The IPCDC helped me understand the situation better.
But this shared vision had not emerged from all the health care sector actors. As mentioned, during the first phases, representatives of the medical profession did not participate in this initiative. We have to mention that the noninvolvement of physicians has been voluntary, to prevent the medical logic, which is centered on offering medical services, from hampering the implementation of a population-based approach, which acts on all determinants of health. Moreover, third sector organizations, and actors outside the health care sector, have not been invited to play a role. This raises questions given that the population-based approach is meant to act on all determinants of health, and therefore would require the involvement of sectors other than health care.
Beyond the initiative’s ability to help members acting at the strategic and system level to better understand the issues related to the reform, certain activities deployed by the initiative, such as training sessions, are designed to promote collective learning within the network, at the local level. The evolution of the formula used in this training, which entailed simultaneously targeting several managers working in the same organization or in the same region, has favored a progressive and collective ownership of the vision underlying the population-based approach. The empirical data indicate that HSSC actors are particularly aware of the importance of two principles of the population-based approach: adapting services to the population’s needs and acting on different determinants of health. On the contrary, it seems that the optimization of resources utilization is less anchored in the HSSC managers’ visions of the population-based approach.
The IPCDC has also successfully facilitated constructive dialogue around the issues of implementing the population-based approach, notably through action guides and training. That being said, the consortium soon realized that the development of boundary knowledge, related to interprofessional and interorganizational collaboration, cannot be achieved through training alone; it requires learning to be grounded in action. Through practice, actors in different organizations and sectors can learn to collaborate and thus develop boundary knowledge.
I profoundly believe that we will learn what can really be done locally in public health, and in population responsibility. But we’ll learn it in the field, by doing it.
The growth of the program to support the HSSCs’ pilot projects aimed explicitly to support this action-based learning.
The evaluation of this program suggests that this latter has produced several learnings regarding (a) the utilization of evidence (b) the health services planning process, and (c) the structuration of collaborative actions.
The IPCDC . . . they bring expertise that is not necessarily present in the field. In the field, we have lots of management expertise . . . but having that expertise . . . population health responsibility . . . , it is like . . . the role of communities, how to establish partnerships, because it is new competencies that they are demanding when they talk about population health responsibility . . . So it takes different forms of knowledge . . . competencies, ways of being different when you act in partnership.
Although this support activity has contributed to learning, its impact is limited, given the IPCDC’s resources.
It’s an organization with all kinds of tentacles, a vision . . . an orientation. . . . It’s not etched in stone, ossified. . . . There is potential for innovation. But the team is so small! . . . .Small team, not much money.
In the opinion of some stakeholders, the orchestration of knowledge capacities requires a more distributed leadership, involving all levels of governance, especially regional agencies.
Getting agencies more involved and coaching is expanding, we are closely linked to the local [health and social services centers]. Agencies follow but are not yet, in my view, proactive or supporting. And if they come help you, all the better. If they don’t cause trouble for you, all the better. But I think that agencies can contribute more to the IPCDC project essentially to support the local [centers].
Ongoing strategizing to frame and direct continuous innovation
This second process is enabled by the development of a concerted, evolving strategy for the actors concerned and by mutual long-term commitment. How does the IPCDC create these dynamics?
During the study period, the impact of the IPCDC on the symposium members’ strategy is observed mainly within the consortium. First, a common strategy to support the transformation of the system has been formulated, which gradually reflected the involvement of all consortium members.
The creation of the IPCDC consortium implied that population responsibility is not the responsibility of the NIPH or of the Ministry. It’s the responsibility of a consortium in which many actors agree to work together.
The implementation of this common strategy transformed the role of some actors. The NIPH is gradually supplementing its usual roles as advisor to the Ministry and hub of expertise in public health by acting as an agent of change in the network. Similarly, RHAs gradually developed their coaching role, going beyond their position of authority figures. This clearly shows the maneuvering room allowed within the IPCDC. Even the Ministry, which did not initially deploy much effort to support the implementation of the reform, included support for the exercise of population responsibility in its 2011 strategic plan (MSSS, 2011):
For all the partners, they all realized that if they go work in the region, as part of a consortium, it gives them a very different image than if they come in isolation. For the Ministry and the regions . . . it is something that lets them [recognize] that . . . they have to adjust, to act outside of their positions of hierarchical authority.
Our data analysis demonstrates that the strategy has evolved to respond to interactions between the actors. The strategy of supporting transformation was gradually oriented toward more systemic actions, which did not simply emphasize spreading the public health vision, as it did initially. Flexibility and dynamism were also observed in the allocation of roles among system actors:
An HSSC asked an agency to develop indicators to measure how well it achieved its responsibility. Right away, the Agency said, “We have to deal with the IPCDC” . . . The IPCDC helped develop this tool.
To summarize, the launch of the IPCDC engendered a strategy of supporting transformation. Yet the dynamics seem fragile. Although the Ministry continues, during this period, to finance the initiative and formally support the implementation of the reform through its strategic plan (cf. 2011 plan), some actors feel that the implementation of the population-based approach is not a high enough priority:
The problem is that the population-based approach, it’s not brought to high places! There was the law, the creation of HSSCs, there were two principles . . . In the law, the hierarchy of services, then the population-based approach . . . But . . . no one said what’s that? . . . . So after that, once it was done, they created HSSCs, and you’re responsible for a population and go ahead . . . The Ministry continued to act the way it always did. And it’s like they don’t give a damn about the population-based approach, population responsibility.
At the local level, during the first years, the IPCDC’s impact on HSSC strategy seems fairly limited.
It’s . . . great awareness of this question of having a population-based approach . . . I relived discussions that we had in our management committee where I was at the HSSC and . . . it’s incredible how that question, the concern for the population wasn’t there. It didn’t come up in our discussions. Okay, it’s normal because . . . Anyway, it’s been three years so the HSSCs were still new. The structures need time to transform. It takes time, a reform.
In fact, the reform has been implemented through the creation of new organizations. These structure changes have monopolized the attention of managers and delayed the work on processes:
And so we found it important at the time to join together because the concern with the population approach was actually something new. Today, it is controlled better in establishments. But at that time, everybody was restructuring their organization. And were maybe less immediately concerned with the population-based approach.
Furthermore, the ongoing incentives (management agreements centering on service volume and accessibility, and financing modes of HSSCs) and daily pressure to produce services do not favor the integration of the new logic of action.
How . . . to raise a challenge of population health responsibility when you manage health services . . . You’re responsible for your people . . . elderly, for your children in difficulty, for this, for that. How do you manage to . . . find their profile, follow this and integrate a population based health approach through it. You often set things aside. Probably because our own lines of conduct, our . . . management agreements . . . are concentrated on management . . . of volumes and we expect to . . . achieve this objective.
Finally, the data suggest that coordination problems inside the Health Department, which affect negatively the consistency of the orientations, are harmful for the implementation of the reform:
Orientations should be more integrated . . . And that, managers, in our work, local managers . . . told us more often than not. The Ministry functions by programs, by branch, and sends many orders down through the silo and branch. So as a result the establishments, everything they can do or the regions is to respond to the orders from the Ministry, and they’ve got little maneuvering room, few possibilities to renew themselves and ask questions, to go forward, etc. So that . . . we have to be clearer, to integrate our orientations.
Developing Public Policy to Embrace Ambiguity
As shown above, the leadership of some entrepreneurs, supported by NIPH, was instrumental to launching the IPCDC. Different health care system actors have been involved over time, including the Ministry. The implementation of a population-based approach is gradually being integrated in public policies, as for instance the 2011 strategic plan of the Ministry aims to reinforce the coherence of actions and to support practice changes. However, it is difficult to consider the ambiguity underlying innovation processes, as illustrated by the pressure exerted on the IPCDC to demonstrate benefits for all partners over the fairly short term:
If we want to be able to interest structures, resources, the Association, the Ministry or agencies, we have to be able to demonstrate concrete actions made that will make people say “we want this” . . . To be able to run this machine, get some bucks, you have to be able to show its value beyond doubt, and I’m not 100% convinced.
Such “resistance” to embracing ambiguity is partly in response to constraints on accountability.
Discussion and Conclusion
As stated by several authors (Hartley et al., 2013; Sorensen & Torfing, 2011; Torfing, 2013), we need to conduct a broad range of empirical studies to better understand how collaborative innovation emerges in the public sector. Our work enriches significantly our understanding. In fact, the analysis of the implementation of the IPCDC allowed us to make two contributions to the state of knowledge, more specifically related to the model proposed by Dougherty and Dunne (2011). To clarify this contribution, we asked two interrelated questions:
How effectively do the activities proposed by the model of Dougherty and Dunne (2011), namely, the orchestration of knowledge capabilities, strategizing and the embracement of ambiguity, favor the emergence of innovations?
Concerning the first question, our analysis suggests that beyond the three activities proposed by Dougherty and Dunne (2011), the emergence of ecologies of innovations demands boundary work to ensure that the actors perceive their mutual interdependence. Our study highlights, in the case examined, advances in this boundary work and positive effects in terms of innovation, which resulted notably from networking between actors. However, our empirical results also demonstrate that the lack of involvement of physicians and of actors outside of the health care sector (acting in the municipal sector, education, etc.) limited the impact of the initiative, notably in terms of collective learning and adaptation of the strategy. The same is true for the lack of integration of orientations issued by various ministry departments, hence the value of boundary work. This work is particularly important in pluralistic environments (Denis et al., 2007) such as the health care sector, where values are multiple or ambiguous, power is shared among diverse stakeholders and where production of services rests on distributed knowledge. The perception of interdependence in these settings is far from a given; we believe that the model of Dougherty and Dunne has neglected this reality.
To support the emergence of ecologies of innovations more effectively, the findings of our study suggest that boundary work particularly requires the deployment of two strategies: (a) strategies to manage cross-boundary connections and (b) boundary breaching, to favor physicians’ commitment.
As part of the first strategy, boundary objects may be very useful to develop collaboration across boundaries. Boundary objects are defined as objects that cross the boundaries between multiple social worlds, used within and adapted to many worlds simultaneously (Star & Griesemer, 1989), and situated amid a group of actors with divergent viewpoints. They adapt to local needs within a social world yet are robust enough to maintain a common identity across sites. We may suggest that devising performance indicators that reflect the results of collective actions, through boundaries (playing the role of boundary objects), may be an interesting lever for managing interdependence among actors, at both the intraorganizational level (e.g., between different branches of a ministry) and at the interorganizational and intersectoral levels (between different sectors of public action). This lever is intended to act on all determinants of health. Beyond the use of these coordination tools, development of intersectoral collaboration would require convincing more actors, outside the health care sector, to contribute to implementing reform and to deploying new forms of governance. To this end, particular efforts to theorize change must be made (Munir, 2005). Theorization of change refers to the development and specification of abstract and general categories, and the formulation of cause and effect chains (Greenwood, Suddaby, & Hinings, 2002), to frame problems (affirm the need for change) and justify innovation (demonstrate the value added contribution). To be effective, this theorization requires the establishment of links between the theory of change and stakeholders’ routines and values (Hargadon & Douglas, 2001; Suddaby & Greenwood, 2005). Future research could clarify concrete means of achieving this goal.
Within the second strategy (boundary breaching), the challenge is to change the dynamics of the medical profession, which, historically, strove through several reform projects to maintain its domination and that displayed much resistance to change, by protecting its autonomy, and prestige, and by controlling resources (Currie, Lockett, Finn, Martin, & Waring, 2012). The mobilization strategy of these actors cannot ignore the prevailing power relations. Some studies suggest that the engagement of these actors and their participation in transforming systems entails particular modes of exercising power, specifically pastoral power (Bejerot & Hasselbladh, 2011):
This subtle form of power invites and encourages subjects to participate in building the structures that govern their conduct. They are expected to assume responsibilities without being ordered to do so, to participate in evaluating their previous actions and to never cease to improve. Thus the freedom enjoyed under pastoral power is in many respect very real but it is also a particular and restricted type of freedom not intended to promote individual and collective autonomy. (p. 1608)
In the Swedish context (Bejerot & Hasselbladh, 2011), physicians’ engagement has occurred in a series of small steps, notably using persuasion, incentives, and successful appeals for almost two decades. In this case, the project initiators simply needed at the beginning to convince some physicians to participate to a quality control initiative, through the utilization of research registers. Gradually, the medical profession has accepted to share, with managers, the responsibility of the modernization of the health care system. This innovation in governance has altered power relationships: From now, the physicians become accountable to the managers. This study suggests that the particular context of the Swedish health care system (there is no particular tradition within the medical profession to articulate their professional position) has facilitated change. Future research should be devoted to the study of the processes of physicians’ engagement in other contexts.
What Are the Levers That Can Foster the Orchestration of Knowledge Capabilities, Strategizing, and the Embracement of Ambiguity?
Concerning our second research question, our empirical analysis of the model proposed by Dougherty and Dunne (2011) shows that several levers contribute to the emergence of ecologies of innovation in highly institutionalized sectors, namely, (a) the role of institutional entrepreneurs who bring together various sources of legitimacy, who strive notably to mobilize several stakeholders and who lead them to collaborate around common issues; these institutional entrepreneurs also play a major role in procuring resources (financial, human, etc.) required for innovation; (b) knowledge mobilization activities (training, think tanks, etc.) that foster development of a common vision of the subject of change and the challenges of its implementation, and that contribute to the theorization of change. The efficiency of these knowledge mobilization activities, in terms of orchestrating learning process, apparently entails targeting actors at various levels of governance (central, regional, local), and the grounding of these activities in territorial realities (i.e., by ensuring that the actors in the same health and social services center (CSSS) and local network organizations participate in the same training); (c) dialogue and decision structures with diverse makeups that allow coordination of collective action. Our study suggests that the fact that these structures are situated on the periphery of the system (outside the “normal” hierarchy) grants the actors maneuvering room and lets them assume new roles (regarding regional agencies, for example); (d) experimental projects at the local level, built with actors in the field based on their needs; this approach favors greater ownership of the change process and allows the actors to learn by trial and error; and (e) coaching activities offered to local actors to support their learning and align local actions with the objectives of the reform.
Collectively, these five levers favor institutional change by supporting different types of institutional work that managers must implement within a reform, particularly structural, cognitive, operational, and relational work (Cloutier, Denis, Langley, & Lamothe, 2016, p. 17). Structural work refers to managerial efforts to establish formalized roles, rule systems, organizing principles, and resources allocation models that support a new policy framework. Conceptual work refers to efforts by managers to establish new beliefs, norms, and interpretive schemes consistent with the new policy. Operational work refers to managerial efforts to implement concrete actions affecting the everyday behaviors of front line professionals that are directly linked with the new policy. Finally, relational work underpins the other three refers to efforts aimed at building linkages, trust, and collaboration between people involved in reform implementation. Cloutier et al. (2016) suggest that it is the interaction and mutual reinforcement of these different types of institutional work that determine their efficacy. In fact, structural work creates agents to engage in conceptual work, operational work and relational work. The conceptual work inspires and partly frames operational and structural work. Operational work tests realism of conceptual work and informs future structural work and finally relational work facilitates structural, conceptual, and operational work. Through its various activities, the IPCDC seeks to equip managers in these various aspects. The training offered to a group of managers within in the same local network can thus support cognitive and relational work. The tools proposed, such as the framework of competencies, are also useful for structural work. Finally, coaching in experimental projects facilitates operational work.
Therefore, our case study confirms that institutional work and distributed leadership may represent levers of innovation. Such work consists in establishing experimental boundaries that are protected from institutional discipline and inventing new practices (Zietsma & Lawrence, 2010), whereas distributed entrepreneurship (Wijen & Ansari, 2006) notably acts on stakeholder mobilization and on the legitimization of change (Lounsbury & Crumley, 2007).
Nonetheless, our study also highlighted the limitations of these levers, notably in terms of adaptation of organizational strategies to the objectives of the reform and concerning consideration of the ambiguity inherent in innovation processes. Several factors explain these limitations. First, although reform is ongoing, some strategic actors do not perceive pressure to radically change practices, unlike in crisis situations (Zietsma & Lawrence, 2010). The IPCDC is struggling to implement an action program independent of the government’s political agenda to raise awareness of the need for change. Second, the implementation of an exploration process, that is, experimentation with new alternatives, in settings that lack a dedicated structure, such as a research and development function, is difficult because the same actors must simultaneously invest in exploration and exploitation processes (March, 1991), which requires contextual ambidexterity (Gibson & Birkinshaw, 2004). In contexts marked by strong pressure to produce, such as the health care sector, the simultaneous implementation of exploration and exploitation processes depends even more on leaders acting at a strategic level who will help develop an environment conducive to ambidexterity (Gibson & Birkinshaw, 2004). The IPCDC strived to act as a leader in these exploration processes, in a context that does not always support these processes. Third, if the marginal (in parallel to the normal hierarchy) position of spaces for innovation can be advantageous because it grants some actors maneuvering room, it raises specific challenges, as these spaces for innovation must evolve in an unchanged environment, characterized by the same operating rules, notably the rules of accountability. Therefore, in the case studied, maintaining a logic of accountability centered on controlling production volumes hinders innovation and the implementation of a reform. Similarly, this logic of accountability, which seeks to minimize the risks inherent in decisions made in the Public Administration, implies that the actors find it difficult to accept and manage the risks related to innovation.
Finally, our study confirmed once again that innovation in sectors where some actors have privileged positions of power, as is the case in the health care sector, is particularly difficult because the sociopolitical context and the resource distribution conditions remain fundamentally unchanged. This finding underlines the need to reflect on ways to exert countervailing power in the system.
In conclusion, this study raises the following question: To what extent can learning processes induced by the creation of spaces for innovation gradually change power relations and ultimately give rise to radical innovations? This question is certainly worth investigating.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research has been financed by a team grant from the Canadian Institutes of Health Research.