Venue Coupling and Actor Circulation in Deliberative Systems: Health Care Governance in Ontario

The “systemic turn” has been one of the most important developments in deliberative democracy in the past decade. Through a deliberative systems approach, scholars and practitioners are challenged to think about ways in which various venues and institutions interact together to produce a healthy democratic subsystem. One major challenge to this approach, however, is its methodological weakness. How exactly are various venues and institutions connected? How do they interact with each other? What conceptual tools are available in making sense of the deliberative system? This article proposes the use of “venue coupling” and “actor circulation” to operationalize some of the key concepts of the deliberative system. Through the case of the Local Health Integration Networks in Ontario, Canada, this article maps the governance system, its institutional and interpersonal components, and their interconnections (or lack thereof). By drawing together key concepts in deliberative democracy and network governance, this article sketches out a framework that can be used to analyze governance contexts in which deliberative practices are fused with traditional political institutions like legislative bodies and bureaucracies.


Introduction
This article examines Local Health Integration Networks (LHINs) in Ontario, Canada through the lens of deliberative governance after the "systemic turn" (Dryzek, 2010). LHINs are regional health authorities with delegated power from the provincial government to make decisions on publicly funded health care investments with mandated stakeholder and citizen participatory mechanisms. This article has two primary goals: (1) to operationalize the key normative concepts of "venue coupling" and "actor circulation" theorized in recent years in relation to each other, and (2) to contemplate the implications for democratic innovations in practice.
Deliberative system theorists maintain that complex governance systems that have various venues, accountabilities, and linkages must be analyzed as a whole, not merely individually, in order to assess their democratic merits (Mansbridge et al., 2012). This approach enables scholars and practitioners to think about decisions being taken in the context of a variety of venues and institutions, interacting together to produce a healthy democratic subsystem. The contributions of this article are primarily empirical and conceptual: a mapping of this governance system and its institutional and interpersonal components, and their interconnections (or lack thereof), as an illustration of the value of the related, but distinct, concepts of venue coupling and actor circulation.
This article begins by describing LHINs in Ontario as an instructive example of broader shifts in public administration and management that Mark Warren (2009, p. 3) has described as "governance-driven democratization," whereby bureaucracies and governance networks can be important sites of democratic reform. Following that, key concepts within the deliberative democracy and network governance literature are drawn together to sketch out a framework that can be used to analyze governance contexts in which deliberative practices in governance are fused with traditional political institutions like legislative bodies and bureaucracies. The case study of LHINs illustrates the value of this approach in analyzing the democratic attributes of system components as an interactive and dynamic whole, but also reveals the challenges of doing so in practice, as the pieces do not fit perfectly together in combination like a jigsaw puzzle as some normative deliberative theorists implicitly contend. However, a deliberative systems approach that foregrounds the analysis of venue coupling and actor circulation can reveal democratic deficiencies in a governance context by focusing on the relationship between various venues and actors.

Local Health Integration Networks in Ontario
Ontario's Local Health Integration Networks were introduced in 2004 to take responsibility for the delivery of services at the local level. The government defined two primary goals: (1) to develop a more integrated and coordinated health care system, bridging hospitals, doctors, home care providers, and long term care facilities to improve the patient experience; and (2) to provide more opportunities for citizens and stakeholders to influence healthcare policy and investment decisions at a more local level. It is the latter part of the mandate that is under examination in this article, but there is evidence that the more participatory and deliberative governance framework established has had a demonstrable effect on inclusive policy reform: the codesigning of culturally appropriate services and supports, or how they brought context and experience to drive reforms in telemedicine, caregiver services for seniors, and multicultural health navigation systems (Doberstein, 2020).
Fourteen LHINs were created across the province in 2004, each with a board of up to 12 members, all appointed by the governing cabinet (though based in part on "advice" from the community), who are expected to have expertise, experience, and an understanding of local health issues and needs, as they collectively fund approximately $30 billion (CAD) in health services to Ontarians. LHINs are arm's-length agencies with accountability links to the provincial government, but with mandated participatory decision-making avenues for stakeholders and communities, and thus also involve accountabilities to the public. LHINs represent an ideal case to examine from a "deliberative systems" approach, which is attentive to the relationships among the various empowered venues and actors, and in particular, how they are connected and iteratively influence each other in a complex system, as captured by the concepts of venue coupling and actor circulation.

Broader Governance Trends
Public administration and management across many democracies in recent years has trended toward more inclusive and collaborative policy-making and implementation involving civil society, and even lay citizens, via networks. Network governance scholars like Eric-Hans Kljin and Joop Koopenjan (2000), among others, have noted this shift toward a bureaucracy that is more open and penetrable to public view, interdependent with civil society for successful implementation, and increasingly charged with intensive public engagement and deliberation. Alex Turrini, Daniela Cristofoli, Francesca Frosini, and Greta Nasi (2010, p. 528) go as far to say that internationally, "in the public sector, the implementation and management of public programmes through networks has now become more the rule than the exception." This trend has also been identified by political theorists in the deliberative democracy tradition, such as Mark Warren (2009, p. 3), who has described this phenomenon of "empowered participation, focused deliberation, and attentiveness to those affected by decisions" within bureaucratic policy-making and administration as governance-driven democratization.
Eva Sørensen and Jacob Torfing (2007) are public management scholars who have identified the democratic implications of such governance trends, though it has not been a primary focus of this literature. Yet democratic theorists have emphasized both the possibilities and limitations of governance-driven democratization. In terms of possibilities, some suggest that these trends increase the chances that "those potentially affected by collective decisions can influence those decisions," a key normative ideal of deliberative democrats (Warren, 2009, p. 3). The rapid growth of venues such as citizens' juries, citizens' assemblies, consensus conferences, online dialogues, deliberative planning, and participatory budgeting-all of which are present in the LHIN governance universe examined in this article-speak to the dramatic changes within modern public administration that are opening up this previously comparatively closed environment, both prior to will formation and after it, especially on how decisions are interpreted and implemented (Parkinson, 2006;Fung, 2003;Boswell, 2016).
While structurally these governance trends are suggestive of democratic possibilities, scholars are also cognizant of its limitations. First among them is that these new venues tend to be elite dominated, not only among the participants (i.e., certain demographics have time, interest, and ability to participate), but also by facilitators and consultants who have built up a cottage industry around public engagement. To Yannis Papadopoulos (2012), these trends look more like stakeholderism or advocacy democracy, rather than as a genuine opportunity for lay citizens to participate in governing for issues that directly affect them. Likewise, Carolyn Hendriks (2008) warns of an environment dominated by industry and government elites at the expense of broader democratic engagement. This is all to suggest that it ought to be a critical line of inquiry in governance studies to examine the democratic character of these arrangements.
Further, it is naïve to think that professionals and experts in a policy subsystem (e.g., health care, in the case of this article) would be willing to exchange and deliberate with ordinary citizens on equal footing as per deliberative norms, and vice versa that such citizens would feel empowered to challenge the authority of such experts. We may normatively prescribe equality of influence, but this does not wash away the potential influence of status perceptions and practices of deference in these contexts. Likewise, John Boswell (2016, p. 724) warns that deliberative democrats often reveal a blind spot for the possible pathologies when it comes to policy implementation, that is, how decisions and policy directions are interpreted and operationalized after will formation, and in particular the ability of powerful actors to "wriggle away from costly actions." Observers of governance-driven democratization are thus careful to emphasize that these trends offer possibilities to inject more citizen participation, representation, and deliberation while also emphasizing that we must critically evaluate how venues are designed and operate in practice to evaluate their democratic character.
The possibilities and limitations presented by governance-driven democratization connect clearly to considerations underlying the so-called systemic approach to deliberative democracy, championed by Jane Mansbridge and her colleagues (2012) in recent years. With the sustained uptake of mini-publics, citizens' juries, and other deliberative forums across governments around the world, critics identify the small-scale nature of these initiatives as representing a fundamental barrier to the integration of democratic deliberation principles into large-scale democratic political systems. That is, for example in LHINs, as we will see, there are many opportunities for citizens to become involved in various types of consultation, engagement, deliberation, and even decisionmaking, and this is enshrined and mandated in legislation. However, we are still only talking about a very small fraction of the total population, and one often not representative of the broader population. This is the "scaling up" problem in the deliberative literature. Deliberative democracy today remains limited to a small number of citizens and thus lacks legitimacy to the broader public, and is too often disconnected from authoritative decision-making. Warren (2007) has argued that we ought to think about how to connect several different types of institutions operating at different points in time and space in the political system, to generate deliberative democratic legitimacy. The importance of the design of venues is connected to a well-established concept known as metagovernance in public administration and management literatures, first popularized by Bob Jessop (2002). The concept of meta-governance is helpful because it captures the relationship and tension between the willingness of the state to engage with stakeholders and citizens in substantive policy planning and decision-making, while maintaining some degree of control over their activity to be consistent with traditional notions of electoral-democratic accountability (Doberstein, 2016a(Doberstein, , 2016b. Meta-governance is about all the choices that need to be made regarding how much authority from the Minister is delegated, and to which bodies, how those bodies are constituted, the processes through which stakeholder and public engagement and deliberation is structured, and how accountability is expressed and rendered (Doberstein, 2013;Doberstein & Millar, 2014). There is broad agreement, however, that whatever choices are made, engagement with stakeholders and citizens ought to be authentic, inclusive, consequential, and linked (in some form) into decision-making (Dryzek, 2007). Hendriks (2008) has further advocated to think about forms of meta-governance that promote the democratic capacity of networks, including anchoring networks to elected officials, making elite institutions more diverse, and monitoring the substantive equality of outputs.
Yet the importance of meta-governance as a practice does not imply that the state should make these determinations on its own, as that may simply reproduce exclusionary politics and governance that stem from electoral democratic institutions. Claudia Landwehr (2015, p. 38) argues for democratic meta-deliberation, arguing that "the legitimation of delegated decisionmaking is not possible without a culture and practice of democratic meta-deliberation which enables reflective institutional design." Dennis Thompson (2008) similarly advocates for a "metadeliberative process," involving a wider public, including sectoral experts, designed to justify how the institutions of governance are structured. Rikki Dean (2017, p. 14), in his typology of public participation, likewise conceptualizes a role for citizens as an "impartial critic of state activity (oversight)." So, while most public management meta-governance scholars point to the centrality of the state as "meta-governors," democratic deliberation theorists invite us to consider broadening the institutional design task to include more actors than those within the state. This suggests that meta-governance design choices themselves must be challengeable and revisable. Metagovernance and meta-deliberation concepts are in some respects the opposite sides of the same coin, revealing the productive bridging of public administration and deliberative democracy concepts, respectively, to analyze modern governance institutions and relationships.
There is also an extensive literature on participatory health service reform and patient and public involvement (PPI) that cannot be fully summarized here, but there are several important areas of alignment with meta-governance and deliberative systems literature. First, Julia Abelson and her colleagues (2004) argue that the meta-governors designing the governance framework to include avenues for public engagement must clearly articulate the purpose of the engagement and its link to the larger decision-making process. They should present basic and technical information clearly and honestly, and use procedural rules that promote power sharing and information exchange among participants and decision-makers. Relatedly, Graham Martin, Pam Carter, and Mike Dent (2018) have documented the challenges associated with public involvement in health system transformation efforts, including resistance among the professionals in the sector, recruitment of motivated and representative citizen participants, and confusion over the primary function of public involvement. In a study of deliberative opportunities in UK's National Health Service (NHS), John Parkinson (2004, p. 391) discovered a pattern whereby public managers tend to separate the "'ordinary people' from the knowledgeable" in venues, which diminishes the potential of information exchange and policy learning via engagement. Finally, Ellen Stewart (2016) differentiates the types of participation among the public in public health system governance into invited realms (e.g., committees, working groups) and uninvited ones (e.g., protests, other forms of resistance), each having different expectations and responsibilities. In this way, the work of health policy scholars researching public involvement connects clearly to the normative basis for venue coupling and actor circulation in the deliberative systems literature, which is introduced below.

Venue Coupling and Actor Circulation
Deliberative systems have two foundational assumptions: (1) that venues needed to be thoughtfully "coupled" and (2) that system design must ensure that influence is structured in a balanced way (Mansbridge et al., 2012). "Coupled" in this context refers to how venues in the system are linked, either institutionally (i.e., the resolutions from one feed into the other part) or relationally (i.e., individuals participating in both sites carrying the outputs or insights from one site to the other, and vice versa) to promote mutual adjustment, learning, and convergence (Hendriks, 2016). Ideally, we would see circulation (of actors) and venue coupling (institutional links) across venues. Nicole Curato and Marit Böker (2016) have thus envisioned "co-development" among different system venues-whether mini-publics, agency boards, working groups, or legislative committees-such that there is a "dialogue across deliberative exchanges," as characterized by Andrew Knops (2016, p. 308). To Curato and Böker (2016), without mechanisms of codevelopment with other arenas, these elements lack external legitimacy if relevant discourses remain trapped inside certain arenas and venues, unchallenged by others and the broader public sphere. There is also concern, advanced by Boswell (2016), that as contingent agreements are made in one venue or at one point in time, their refinement and implementation, in particular, is not limited to particular actors or interests who have resources or expertise to stick with it. Instead, there needs to be opportunities institutionalized through the policy process for all. Scrutiny forums, contestatory reviews, and feedback funnels are mechanisms to reduce the possibility of powerful actors to manipulate activities downstream in the policy process to their preferred ends (Boswell, 2016). Thus, actor circulation and venue coupling are critical pieces of any democratic subsystem, as these are institutional and interpersonal mechanisms to avoid the pathologies of implementation.
As much as it is generally accepted that venue coupling and actor circulation are essential within a deliberative system, the concepts need to be further operationalized in order to more systematically identify and measure their presence or absence in these contexts. Curato and Böker (2016, p. 186), while advancing the literature by examining the various legitimacy dimensions of mini-publics in relation to the broader democratic system, do not explore deeply what precisely are "mutually productive interactions between different components and functions of deliberative systems." Likewise, Hendriks (2016), although clear on the need for "designed coupling" given that it is unlikely to happen naturally and offers several examples of desirable coupling that links citizens and parliaments, devotes less attention to conceptualizing a general typology of venue coupling and actor circulation. Finally, Ricardo Mendonça (2013, p. 15) emphasizes the importance of individuals "criss-crossing informal and formal settings… linking the diverse moments of a broad deliberative process" but did not categorize the various ways in which this might occur. As such, there is an opportunity to build on the conceptual work initiated by these scholars who have focused on the "connective tissue" between venues in democratic systems.
Venue coupling and actor circulation are related but distinct concepts. In this vein, I suggest that venue coupling be reserved as a concept that captures the various institutional links that may be created in a policy subsystem. That may include the following, which are implied, though not clearly specified and differentiated, in the existing literature: (1) a reporting linkage, whereby one venue issues information reports to another; (2) a recommendation linkage, whereby one venue issues recommendations to be considered at another decision-making venue; and (3) an implementation linkage, whereby one venue makes decisions that must be interpreted and implemented by another.
Actor circulation, on the other hand, ought to refer to the interpersonal linkages across venues, and may include the following: (1) shared membership across venues, whereby some actors are participants in multiple venues in the system; (2) guest membership across venues, whereby some actors visit venues for a specific purpose; and (3) a public engagement broker, an individual whose primary task is to share developments across the policy subsystem. Though they are distinct in these ways, both venue coupling and actor circulation capture how information, perspectives, expertise, proposals, recommendations, and decisions travel through a policy subsystem as part of the co-development of policy planning and implementation.

Methodology
The case of LHINs in Ontario demonstrates the operationalization and measurement of normative concepts emerging from the deliberative systems literature. By examining how venue coupling and actor circulation manifest in practice, we can evaluate their alignment with normative claims. LHINs also provide unique analytical opportunities because there are 14 of them across the province of Ontario. While they all exist in an identical political-legal context, they are permitted to vary in terms of governance and democratic mechanisms according to local preferences and conditions. As such, they exhibit variation that can be utilized to examine the democratic character of governance patterns in an otherwise controlled environment.
In mapping and analyzing various governance bodies and their relationships, the LHIN universe requires a multimethod approach that draws on data over a period of time. I present findings for the LHINs by drawing on archival records including legislation, policy documents, annual reports, consultation documents, and Hansard debates and committee hearings. Key legislative documents include the Local Health System Integration Act (2006) and the Patients First Act (2016), and policy documents like the accountability agreements between the Ministry of Health and each LHIN. Other important documents analyzed from the 14 LHINs include: annual reports, strategic plans, business plans, community engagement reports, and minutes of the meeting from LHIN boards and subcommittees. I also conducted interviews with key policy actors; among them are the current and former LHIN board members, key political and bureaucratic officials in the Ministry of Health and Long-term Care, LHIN staff, stakeholders, and citizens who have been involved in advisory committees, citizens' panels, and consultations. 1

Examining LHINs in Ontario through a Deliberative Systems Lens
There is enormous complexity involved in health care governance in Ontario, Canada. For the purposes of this analysis, key actors and institutions include the Minister (of Health) and Ministry bureaucracy, LHIN boards, LHIN stakeholder committees (of which there are dozens per LHIN), citizens' panels as mini-publics, and the broader interested public, each with different responsibilities and expectations, as shown below in a graphical depiction in Figure 1. Each has a distinct role in policy-making and governance, with the Minister (and Cabinet) establishing the overarching policy goals for health care in the province, the LHIN boards deciding how those goals ought to be achieved in the local context, informed from policy analysis and democratic deliberation among stakeholder committees, citizens' panels, and the broader interested public. Stakeholders and citizens are part of both policy development, but also in elements of implementation and program evaluation. Each of these actors and institutions has democratic legitimacy from different sources: ministers and the cabinet from electoral politics; LHIN boards from expertise and health care policy experience; stakeholder advisory committees from interestbased principles; and citizens' panels from their position as citizens, taxpayers, and consumers of health care.
There are formal and informal relationships between the layers of venues in this realm. For example, as presented in Figure 1, the Minister of Health has a meta-governing relationship to LHIN boards by specifying the Memorandum of Understanding (MoU), the Ministry-LHIN Accountability Agreements (MLAA), and the Provincial Strategic Plan (of overarching goals for health care that LHINs must work toward), and ultimately evaluating their performance. In the other direction, LHINs formally report to the Minister on a regular basis, including with their Annual Business Plan (on how they intend to spend the public dollars transferred to them by the Ministry) and the Annual Report (summarizing activities, initiatives and performance). Further down the institutional chain is the relationship between LHINs and their stakeholder and citizen advisory groups, which is mutually influential: the committees and panels provide advice and information to the LHIN board, while stakeholders in the health system are ultimately responsible to help implement the agenda decided by the LHIN. Finally, a governance framework from the normative position of a deliberative system would have dual tracks of influence in its basic design, from the state (meta-governance) and from the stakeholders and citizens (meta-deliberation), as well as transparency and publicity flowing downward from every institution to the interested public. In Figure 1, we also see how venue coupling and actor circulation as normative concepts can become operationalized in the LHIN subsystem, with blue arrows indicating the substantive nature of the coupling link across venues, and the red arrows indicating where actor circulation ought to be observed to promote cross-pollination of ideas and perspectives. Figure 1 specifies what we are looking for in a democratic subsystem, normatively speaking, in terms of venue coupling and actor circulation. Figure 2, meanwhile, provides a summary of what we actually find when we examine most LHINs in practice, and reveals that important venues and actors, while present in the field, are not always meaningfully integrated or even connected, thus revealing venues operating in isolation from each other and thus failing to exhibit the connective tissue that systems theorists advocate. Examining governance contexts in this manner can not only identify democratic shortcomings, but also allow for a reflection on what type of connections we ought to expect across venues. The paragraphs below draw on empirical data collected in LHIN governance context that tells this story of missing connections between venues and actors. Note: Boxes indicate key venues in the LHIN governance system, blue arrows reference elements of venue coupling, and red arrows reference circulation of actors across venues, all as an ideal representation.
Recall that a governance framework from the normative position of "deliberative systems" would have dual tracks of influence in its basic design, from the state (meta-governance) and from stakeholders and citizens (meta-deliberation). When examining the creation of LHINs, the balance tilts heavily, almost entirely, to meta-governance, with very little stakeholder or citizen involvement in the debate or design of the LHIN governance framework when they were created. The idea to create LHINs was subject to little public debate, as it was not presented to the public as part of health system reform in the 2003 election, and its design was principally driven by a small elite "dream team" of advisors brought into the Minister's office. Critical decisions such as the proposed authority to be devolved from the provincial to the local level LHINs, their geographical boundaries, and the appointment process of LHIN board members, among others, were debated and decided internally within the state, with no substantial preconsultation with stakeholders or citizens. And notably, all of this was done by executive order initially, that is, without a legislative foundation that would prompt at least some form of opportunity for stakeholder and public comment. Legislative committee testimony and media coverage reveal that citizens and stakeholders were caught off guard by the sudden introduction of LHINs and found it difficult to comment on them for legislators because they knew very little about what they were. This is critical because from the beginning, LHINs have suffered a legitimacy deficit as birthed from the shadows with no substantive stakeholder or public input into their design or purpose. Colleta McGrath from the Quest Community Health Centre identifies the political importance of meta-deliberation when she says that "No matter how good a new health system paradigm is, for it to work effectively and be supported by residents, people must identify with some component of the paradigm." The overwhelming evidence gathered from legislative testimony and interviews with LHIN and Ministry staff suggest that it was the original process of introduction of LHINs without sufficient meta-deliberation that left major local stakeholders and citizens alienated from the reform effort.
Though the enabling legislation that created LHINs devolved considerable governance and administrative authority from the Ministry, the Minister retains critical meta-governance (oversight and steering) roles with respect to LHINs in terms of setting provincial policy priorities, devising enabling and constraining legislation on LHINs, and acting as the holistic system overseer that encourages LHINs to integrate services within and across their regions. There have always been formal dimensions to the Minister-LHIN relationship, including the MOU, the MLAAs, and the Ministry-approved annual Business Plan for each LHIN, but much of what has been contested by LHINs as overbearing are the less formal, and sometimes arbitrary, interventions to shape LHIN activities. The Minister (and by extension, Ministry officials) at times meddles in LHIN affairs in ways that seem to violate the spirit of the enabling legislation. From arbitrary interventions that in essence overturn LHIN decisions to attempts to reach in and shape LHIN decision-making (e.g., some LHINs report that their staff spend at least 30% of their time responding to Ministry inquiries), the various ministers overseeing LHINs over the years have had a strong presence. Yet the Ministry has also been criticized for, on one hand, not holding LHINs accountable when they fail to meet certain performance targets (i.e., being too hands off), while on the other, for intervening inappropriately in LHIN decision-making (i.e., being too hands on), which can be rather disruptive to LHINs, especially in terms of their legitimacy in their communities when they are vetoed from above. Note: Boxes indicate key venues in the LHIN governance system, blue arrows reference elements of venue coupling, and red arrows reference circulation of actors across venues, all as an ideal representation. An X refers to severed venue coupling and actor circulation in practice.
The LHIN boards, which are the governance entities formally accountable to the Minister of Health, are appointed by the provincial cabinet, a portion of which is upon the recommendation of a community-based process. There is wide diversity of leadership patterns within and across the 14 LHINs in terms of board governance, including issues of inclusiveness, expertise, and connection to their community. Examining the composition of the boards-in particular how they attempt to balance the objectives of a competency-based versus a constituency-based board-we see that in recent years they have leaned heavily toward skills rather than representation. 2 The selection of the board, the exclusive body with direct accountability to the Minister, remains controversial to some who question how their appointment by the Cabinet-rather than election or a broader community-driven process for appointment-renders their accountability relationship to the government, not their community. Yet there are various opportunities in which the community and stakeholders can influence the board through public deputations at board meetings, media oversight, and advocacy pressure. The LHINs are legislated with significant transparency demands, and while they do not always live up to those ideals, many LHINs exhibit active disclosure of their activities and provide many opportunities for citizens and stakeholders to follow their decisions and react accordingly.
The relationships between LHINs, stakeholders, and citizens are part of the governance framework that is less formally specified in law, but nonetheless exists in practice. Public engagement is a widespread practice among the LHINs, but there have been challenges associated with deciding how each piece contributes to the broader goals and activities of LHINs. "Community engagement," as defined by the enabling legislation of LHINs, includes patients and citizens, health service providers, and employees in the health system, and this is primarily achieved through an extensive array of stakeholder and patient advisory committees at each LHIN-the only one required by law as originally written was a Health Professionals Advisory Committee. Physicians in particular made the case that they need formal input… We're the people who directly look after patients… where the responsibility ultimately lies for people, especially when they're in hospital. So, we think we have a lot to contribute as far as advice. In no way do we want to make final decisions, but we think our advice is valuable. 3 One senior LHIN official in charge of community engagement suggested that their LHIN had 25 to 30 advisory committees at any given time, "where we involve patients, families, caregivers, health professionals, and more in the design, development, monitoring of the programs that we are creating." 4 Yet one might contemplate whether citizens have sufficient knowledge and engagement to understand the complexity of the health care sector to provide advice and feedback to LHINs. Another LHIN official suggested that education, citizen feedback, and engagement are essential because "the data [we have in the health care sector] is great… but you can't say what is that community's first order of priority for what we need to do." 5 Undoubtedly, LHINs have historically engaged more systematically with health service providers and advocates than they have with so-called ordinary citizens, but this has changed in recent years with the growth of citizens' panels in various LHINs. There is considerable evidence of representative panels of 10 to 20 members, who are provided small honoraria, as being among the most effective means to involve the public in a sophisticated process of priority setting via deliberative engagement (Dickinson, 2002). For example, the Central LHIN created the Citizen's Health and Advisory Panel in 2013, consisting of nine members randomly selected among hundreds of applicants, which meets four times annually to provide advice to the LHIN on a variety of issues. The panel is subject to regular turnover, as each member is normally appointed to threeyear terms to inject new perspectives while at the same time building up capacity to engage within their term of service. Central LHIN panel members largely report positive perceptions of the process, commenting that they can see their group's work reflected in the business plan and other smaller initiatives advanced or support by the LHIN (Central LHIN, 2015).
But experience with citizens' panels is not universally positive. Toronto Central (TC) LHIN's panelists, for example, when interviewed, report mixed results. One member, Alies Maybee, thought she has had some influence on LHIN decisions, reporting that "some of the ideas I had around sub-region structure for patient engagement were adopted" but struggled to find additional examples of insights emerging from their group that were directly adopted by the LHIN. Peter Cresswell, another member of the TC citizens' panel, also noted that a problem was that "they engaged [with us] after they'd already had their plans together." 6 This is a common problem identified in the citizen engagement literature on the importance of clearly establishing the expectations of these venues, including their expected tasks and the scope of their activity. Another panelist, Aditya Muralidhar, was more direct when asked if she feels like the citizens' panel is making a contribution: "No. I hate being blunt about it, but no. Because nothing I've even said has been undertaken." 7 Where does the feedback from the citizens' panel go, she was asked, and she replied that their points are heard "but then filed away for the national archives. That's about it." Maybee sees it as critical that the LHIN needs to report back to the citizens' panel about how their input was used or not, or where it was channeled, "otherwise people will walk." 8 In this way, Maybee was pleading for what scholars have identified as venue coupling and actor circulation in complex governance contexts.

Designed Coupling: Circulation and Venue Coupling in LHINs
The case of LHIN governance reveals that, generally speaking, the venues and actors are diverse and inclusive in the spirit of the all-affected principle from the deliberative literature. Yet a systems approach demands that we examine the cross-pollination of actors across venues in order to assess the democratic character of the system as a whole. That is, a democratic subsystem must ensure actor circulation and venue coupling across the levels, that is, interpersonal and institutional links across venues, where activities and byproducts from one venue feeds into another.
What tends to happen in LHINs is that actor circulation is rare, though venue coupling is stronger than it may appear at first blush, with various institutionalized linkages forming a ladder of sorts that build out initiatives from the ground level to influence higher-level provincial policy-making. In terms of circulation of actors across venues, some LHINs have had participants in their citizens' panels present findings or outcomes from their work to the respective LHIN board, thus representing actors from one venue bringing lessons to other venues to influence their work. More often the case, however, is that circulation (if it happens at all) takes the form of information, not people, flowing from one venue to the next. Yet circulation of actors is also critical, as evidenced by the overwhelming impression given in interviews with citizens' panel members from the Toronto Central LHIN who did not get this opportunity, and that they had virtually no idea where their work went and what influence it had on "higher-level" decision makers. 9 To Maybee, a citizens' panel member, "we'll have a focus group on X [topic], and then nothing comes back to the focus group and people say 'Well, I never know if it makes any difference or not, so why should I bother.' So, you break trust." 10 By carrying their own ideas and outputs to other venues, rather than relying on LHIN staff, confidence in their work and trust in the process would be enhanced.
Whereas actor circulation is concerned primarily with the movement of people across venues, venue coupling is related, but distinct, in the sense that it is concerned with fostering institutional links across venues, where outputs or insights from one venue feeds into other venues. This could mean recommendations for action being taken up for consideration by a "higher-level" body. It could also mean draft ideas of priority areas for funding are sent "down" to the public for consideration and ranking, or any other mechanism through which the products of one venue are institutionally linked to another, ideally building on the work from the other. For example, from one venue to the next, there would be a coherent development and release of public reports of their activities that are considered and operationalized at those venues. Stakeholder committees may conduct work that feeds into the development of the annual business plan, which is approved by the LHIN board and subsequently by the Minister, for example. Yet we see that citizens' panels are often severed from these institutional linkages, often created and operationalized to help the LHIN set priorities and make decisions but can have participants report few tangible contributions that we reflected in plans that move up the chain of authority. Likewise, citizens' panels have been disconnected from the broader interested public in many cases, with little public information released about their activities, decisions, and outputs among others, in a way that the broader public (or media) could evaluate, as shown in the graphical depiction of venue coupling and actor circulation in Figure 2.
Yet there are various opportunities in which LHINs intersect with the broader interested public, including at the board level, advisory committees, and citizens' panels. At the board level, some LHINs have open mic opportunities at board meetings. Citizens can raise their unfiltered concerns to LHIN leadership, and some have regularized media availability after public board meetings. 9 Peter Cresswell, interview, January 8, 2018; Barbara Fallon (Citizen participant on TC LHIN Citizens' panel), interview, January 6, 2018. 10 Alies Maybee, interview, January 5, 2018.
Others use webinars or webcast technology to allow the public to watch proceedings. 11 Beyond the LHIN board-level interactions with citizens/public space, numerous advisory committees created by LHINs involve stakeholders and citizens. However, these are often focused on the implementation and monitoring of programs, not idea generation or reform. 12 Leaving implementation and monitoring to stakeholder advisory committees also risks selective adoption of policy according to their preferences. This is an area where citizens' panel venue coupling and actor circulation within the work of stakeholder advisory committees would serve an important oversight function.
Indeed, a key lesson that emerges from this evaluation of LHINs through a systemic lens is that citizens' panels need to be more thoughtfully created and linked to other actors and venues in the local context. Too often they are decoupled from much of what is going on around them, rarely feeding into the work of other venues and institutions in this system. A major warning sign for the citizens' panels was the near-universal condemnation of the Toronto Central LHIN by participants interviewed. Part of the virtue of citizens' panels is to use them to enhance the legitimacy of the system and generate confidence among the public for the work being done, and this essentially backfired. Citizens' panels thus need to be designed with institutional linkages to other venues where members circulate across the system. One suggestion offered was to have some overlapping representation on a citizens' panel with health professionals, who often have their own committee or working groups, and to have these bodies intersect from time to time to jointly puzzle through issues. Toronto Central citizens' panel member Muralidhar expressed a desire for some health care professionals to appear before the citizens' panel, as "it would be nice to have someone who knows about medicine to be involved in that process as well," suggestive of actor circulation or coupling being a gap in this LHIN engagement framework. 13 Muralidhar suggested that health professionals ought to engage with their work through presentations and information exchange so that the citizens' panel is not estranged from the broader work of LHIN advisory committees.
In this view, citizen-identified issues in the health system could then be partially explained or put into context by health professionals. Likewise, health professionals can gain a better understanding of the health system from the patient's view. Creating structured opportunities like this, where the various committees within LHINs are brought together for joint planning or deliberations on specific issues, would be a way to further enhance actor circulation and venue coupling. This type of structured cross-pollination is akin to Boswell's feedback funnels, where in this case, citizens representatives reflect on service delivery to stakeholder advisory committees (the members of which are often tasked with the implementation of higher-level decision making in the context of health care), in part to ensure that they have not drifted from the envisioned policy or programmatic direction.

Discussion
This article set forth to operationalize key concepts of venue coupling and actor circulation that emerge from the deliberative systems literature, and evaluate the complex governance relationships captured in health care governance in Ontario. There are two key lessons from LHINs that speak to larger questions about the design and management of deliberative governance institutions in practice.
First among them is that while there is extensive public engagement through various mechanisms, initiated at various times and for different populations (i.e., citizens, stakeholders, patient groups, etc.), there has been a pattern of engagement without clarity of purpose or outcomes. That is, it has the appearance of throwing the kitchen sink at the issue of public engagement, with at times a lack of strategic objectives, and importantly, without enough care to establish feedback loops with those engaged on how their involvement mattered to the process. The most obvious example of this is the uneven rollout of citizens' panels at most LHINs, which, while they are a welcome deviation from the otherwise heavy stakeholder bias in public engagement vis-à-vis LHINs, there has been some patterned dissatisfaction among participants.
A second area of weakness from a systemic point of view relates to venue coupling and actor circulation. It is not always clear how efforts at one level of governance contribute to the developments and debates in other levels, which threaten sustained participation and the substantive gains that can result from engaging different policy actors at different points, for particular purposes. In some ways, outputs from various venues in the broader system have to "go somewhere," and feed into policy making processes up and down the chain of decision-making and implementation for there to be contributions and accountability infused across the system. This can be achieved by inserting formal reporting procedures from one venue to another, but also in terms of the circulation of actors from one venue to another. One recommendation made from several people interviewed was for LHINs to employ a "public engagement broker," whose main task is to travel across venues, inject outputs from them throughout the system, as well as report back to lower-level venues about how their work became embedded in the larger system of planning.
The case of Ontario's LHINs has provided the opportunity to further operationalize the concepts of venue coupling and actor circulation that serve as the "connective tissue" in deliberative systems. Venue coupling and actor circulation, in the formulation advanced here, are related but distinct concepts. They both capture how information, perspectives, expertise, proposals, recommendations, and decisions travel through a policy subsystem. They remain distinct mechanisms in order to do this, with each essential to ensure that the diversity of inputs in a democratic subsystem are reflected in the process and outputs. In this vein, I have specified and differentiated various types of venue coupling and actor circulation, and revealed in the empirical analysis that actors in the LHIN context demand both types of "connective tissue" across their policy subsystem. Table 1 differentiates the general types of venue coupling and actor circulation, with examples of those ideal linkages and relationships in the context of Ontario's LHINs. The gaps across venues in the case of LHINs in Ontario are specific to this context, but there is a broader potential to theorize how, when, and where the various types of venue coupling and actor circulation ought to be activated in a more general sense. Thus, future conceptual work ought to be devoted to pairing ideal types of venue coupling and actor circulation in terms of the common sets of venues in most governance contexts: legislatures, executive departments, quasiindependent boards, stakeholder advisory committees, citizens' panels, and the broader interested public. That is, what are types of venue coupling and actor circulation best suited to fostering the connective tissue from citizens' panels to stakeholder advisory committees? Is it a combination of a recommendation linkage (type of venue coupling) and guest membership (type of actor circulation)? Or would a reporting linkage be more appropriate in combination with a public engagement broker? How does this change when considering fostering connective tissue between quasi-independent boards and the legislatures from whom they are delegated authority? For example, in service of promoting actor circulation in this context, we would not want the local legislative representative to have shared membership such that she is also a member of the quasiindependent local board (as that would compromise its quasi-independence), but it would likely be quite valuable to have that elected official be engaged as a guest in that venue from time to time to transmit perspectives and pressures from the political realm. Further theorizing these dimensions, while beyond the scope of this article, ought to be an important area for future work.

Conclusion
The primary lesson on venue coupling and actor circulation, key concepts advanced in this analysis, is that when there are a plurality of legitimate actors and institutions working in a complex governance environment, there must be clarity on their particular role in the division of labor and structured linkages between them. Where there can be an enormous number of islands in complex policy subsystems, the key task of governance is to devise an efficient and responsive ferry system to connect the islands, so to speak. That is, venues must be coupled, meaning that outputs from one feed to another, either in the form of reporting (i.e., this is what we discussed and the position/s we arrived at) or formal recommendations that must be considered by the other venues. Institutional linkages between venues must also be reinforced by the circulation of actors within the system. This can involve a member of a citizens' panel appearing at a stakeholder committee, or even a legislative committee hearing, to provide the connection between the work conducted in particular venues to the other parts of the system. This does not imply a deterministic "use" of deliberations, recommendations, or decisions, but rather that the outputs from venues are shared, considered, and built upon as decisions make their way through a complex governance system. This is what governance-driven democratization looks like: it can look muddled, at times be frustrating for those involved, and often lacks a direct relationship in terms of inputs to outputs that some observers demand. But one thing is clear: this type of governance arrangement is fundamentally more democratic than previous systems of health care governance in Ontario. There have never been more opportunities for citizens, stakeholders, experts, and the like to be involved in shaping health care planning and decision-making. And while at times it can appear messy or fuzzy, it is more responsive and inclusive to diverse perspectives and input than previous governance models.
Getting the critique right is the first step to contemplating reforms that may help improve the structure and function of deliberative governance spaces. A systems-analytic approach at the subsystem level can help evaluate the component parts and how they are (or are not) connected to each other to present a more sophisticated critique of the democratic character of the governance context.