Enhancing inter-organisational partnerships in integrated care models for older adults: a multiple case study

Purpose – The purpose of this paper was to develop deeper insights into the practices enacted by entrepreneurial healthcare managers to enhance the implementation of a partnership logic in integrated care models for older adults. Design/methodology/approach – A multiple case study design in two urban centres in two jurisdictions in Canada, Ontario and Quebec. Data collection included 65 semi-structured interviews with policymakers, managersandprovidersandanalysisofkeypolicydocuments.Theinstitutionalentrepreneurtheoryprovidedthetheoreticallensandinformedareflexiveiterativedataanalysis. Findings – While each case faced unique challenges, there were similarities and differences in how managers enhanced a partnership ’ s institutional logic. In both cases, entrepreneurial healthcare managers created new roles,negotiatedmutuallybeneficialagreementsandco-locatedstafftofosterinter-organisationalpartnershipsbetweenpublic,privateandcommunityorganisationsinthecontinuumofcareforolderadults.Inaddition,managersinOntariosecuredadditionalfunding,whilemanagersinQuebecorganisedbiannualmeetingsandjointtrainingtoenhanceinter-organisationalpartnerships. Originality/value – This study has two main implications. First, efforts to enhance inter-organisational partnerships should strategically include institutional entrepreneurs. Second, successful institutional changes maybesupportedbyinvestinginintegratedimplementationstrategiesthattargetrolesofstaff,co-locationandinter-organisationalagreements. There are two different computer systems. It can ’ t work. And then finally we invited CCAC Care Co-ordinators to come to our [Virtual Ward] rounds and they have access to the system. ” It is not entirely clear what the social worker will do in Family Medicine Groups. And even the reception of social workers in Family Medicine Groups is not very enthusiastic for many Family Medicine Groups (Q-Policymaker 005) Our study to by practices enacted by healthcare managers that entrepreneur roles in their efforts to transform their organisations towards more collaborative practices. Our findings show that entrepreneurial managers creatively adjusted the operations of their organisations by introducing new roles, promoting the co-location of staff or negotiating mutual agreements to enhance partnership working. While these organisational strategies are frequently reported the our findings also showed that entrepreneurial actions of managers were critical to overcoming or working around specific challenges in order to support partnerships. For example, healthcare managers creatively used data sharing or service agreements to improve partnership working within the structural constraints of their respective institutional environments.


Introduction
Frail older adults living with multiple chronic diseases often need a broad range of health, social and community-based services to improve health outcomes (B eland and Hollander, 2011;Kodner, 2006;MacAdam, 2011). Over the last two decades, health systems of developed countries have implemented community-based integrated care models as effective and efficient means to address the complex needs of community-dwelling older adults (B eland and Hollander, 2011;Goodwin, 2016;MacAdam, 2011). Integrated care models are complex multi-component interventions that consist of collaborative clinical and administrative practices across acute, primary and community care sectors (Goodwin, 2016;Kodner, 2006;Kodner and Spreeuwenberg, 2002;Lewis et al., 2010;MacAdam, 2011). Interprofessional teams, professional care coordination and shared clinical tools are key features of integrated care models.
Inter-organisational partnerships that ensure seamless service delivery is an important aspect of integrated care models because complex patients often require a mix of services from different provider organisations (Auschra, 2018;Goodwin, 2016;Kodner, 2006;Suter et al., 2009). In the field of integrated care, institutional change needed to enhance interorganisational partnerships often involves building formal and informal relationships between stakeholders, mutual exchange of information and sharing key resources in order to forge stronger links between various health and social care organisations (Auschra, 2018;Cook et al., 2007;Lewis et al., 2010;MacAdam, 2011;Shaw et al., 2017). Previous works show that the configuration and extent of inter-organisational partnerships varies from centralised models (Birrell and Heenan, 2014;Forbes, 2012) to organic partnership initiatives that developed in ways most suited to their local circumstances (Ahgren and Axelsson, 2007;Shaw et al., 2017). For example, Integrated Care Partnerships in Northern Ireland are a centralised model consisting of formal committees of representatives from health and social services organisations of a given health region (Birrell and Heenan, 2014), while Chains of Care are organic partnership models that developed in local councils in Sweden (Ahgren and Axelsson, 2007). While most studies have focused on positive outcomes of interorganisational partnerships in terms of improved access to services, improved quality of service delivery or improved client satisfaction (Goodwin et al., 2012;Kodner, 2006), less attention has been paid to the substantial efforts that are employed to implement these interorganisational collaborative models in real life contexts (Auschra, 2018;Cook et al., 2007;Lewis et al., 2010).
The behaviours and practices of social actors in organisational settings are often driven by institutional logics or widely shared values of their organisational field (Greenwood et al., 2017). A partnership institutional logic, defined as widely shared ideas or values driving an array of strategic actions to bring together two or more groups of social actors to achieve a common purpose (Cook et al., 2007;Forbes, 2012;Kodner, 2006;Nies, 2006;Shaw et al., 2017), represents a paradigm shift from traditional siloed and fragmented care towards more collaborative and continuous models (Auschra, 2018;Goodwin, 2016;Kodner and Spreeuwenberg, 2002). Efforts to implement a partnership logic in integrated care models often face important challenges, such as difficulties in engaging key partners (especially social services) on joint committees (Birrell and Heenan, 2014), difficulties in balancing the interests and goals of various organisations involved in the partnership (Cook et al., 2007), or substantial efforts to convince physicians of the value of partnerships (Forbes, 2012). Healthcare managers play a unique role in shaping efforts to implement partnerships because management decisions represent strategic and operational commitment of organisations to adapt and change their practices (Auschra, 2018;Breton et al., 2014;Cook et al., 2007;Forbes, 2012;Nies, 2006;Shaw et al., 2017). In-depth understanding of effective practices associated with enhancing inter-organisational partnerships is essential to identifying areas to improve the uptake of integrated care policies. Hence, the main objective of this study is to develop deeper insights into the practices enacted by healthcare managers to enhance the implementation of a partnership logic in integrated care models for older adults.
Institutional entrepreneur theory Institutional theory examines how new organisational arrangements emerge under paradigm shifts or the introduction of new institutional ideas and values (DiMaggio and Powell, 1983;Greenwood et al., 2017;Meyer and Rowan, 1977). Institutions can be understood as widely shared beliefs and values that are often embedded in policies that influence the coordinated action of social actors (Greenwood et al., 2017). The traditional institutional approach focused on analysing the role of institutions in shaping the preferences and behaviours of social actors; highlighting increasing conformity and similarity in organisational forms as dominant institutional ideas spread within an organisational field (Greenwood et al., 2017). However, the emergence of the neo-institutional approach put more emphasis on the agency or the independent actions of social actors within the constraints of their institutional environment (Greenwood et al., 2017). A central tenet of the neo-institutional approach is that institutional change occurs when key social actors embrace and shape environmental ideas, beliefs and values that provide motives for action to what is considered appropriate or inappropriate behaviour in their workplace (DiMaggio and Powell, 1983). In the field of integrated care, a partnership institutional logic is a key feature of healthcare policies and practices in several developed countries (Birrell and Heenan, 2014;Forbes, 2012;Kodner, 2006;Shaw et al., 2017). One way of understanding the efforts of social actors to enact collaborative practices that are driven by a partnership logic is through the concept of institutional entrepreneurship (Forbes, 2012;Shaw et al., 2017).
Institutional entrepreneurship is a concept associated with institutional theory referring to the "activities of actors who have an interest in particular institutional arrangements and who leverage resources to create new institutions or to transform existing ones" (Maguire et al., 2004, p. 657). In other words, institutional entrepreneurs are individuals or collective actors who introduce new practices or modify existing practices within the constraints of their institutional environments (Battilana et al., 2009;Garud et al., 2007;Maguire et al., 2004). For instance, healthcare managers may act as collective institutional entrepreneurs when they connect new treatment norms to the routines of providers (Maguire et al., 2004).
The literature suggests that the capacity to act as an institutional entrepreneur is preconditioned by two types of enabling conditions. First, are exogenous factors which are characteristics of the organisational field or the environments of social actors (Battilana et al., 2009;Jensen and Fersch, 2019). Exogenous factors cause disruptions of organisational fields and invite new ideas or different ways of doing things (Battilana et al., 2009). An exogenous factor that might facilitate the emergence of institutional entrepreneurs include regulatory changes, such as the introduction of a new law or policy reforms favouring partnership working between various organisations involved in the continuum of care for older adults (Birrell and Heenan, 2014;Forbes, 2012;Shaw et al., 2017). Another exogenous factor includes a change in leadership that modifies the strategic orientation of an organisation towards more integrated care .

Implementing integrated care for older adults
The second type of enabling condition are endogenous factors, referring to the way social actors perceive their organisational environment and access critical resources necessary to enact changes they prioritise (Battilana et al., 2009;Jensen and Fersch, 2019). Some endogenous factors that might influence the activities of institutional entrepreneurs include: 1) Sense-making: the way social actors give meaning to equivocal informationsuch as laws, regulations, directions -and act accordingly to establish practices they prioritise (Denis et al., 2009;Garud et al., 2007); 2) Social capital: hierarchical position, formal and informal networks influence the likelihood for social actors to build relationships that are necessary to enact changes they prioritize (Battilana et al., 2009). 3) Power: the capacity to mobilise resources and to control the behaviour of individuals to achieve integrated care partnerships (Garud et al., 2007).
Recent studies have illustrated how the institutional entrepreneurship of healthcare managers has contributed to enhancing a partnership's institutional logic in the field of integrated care. Forbes (2012) studied the implementation of a Community Health Partnership Model that was mandated by the Scottish Executive. They showed that as institutional entrepreneurs, healthcare managers transformed this centrally mandated partnership model by engaging in institutional work that involved re-designing the model according to local priorities and then lobbying to legitimise the modified partnership model at the local and system level. Shaw et al. (2017) showed how healthcare managers improved care transitions between the hospital and community sectors in England by creating a hub bringing together local providers to work together and investing in relationship work that facilitated partnership working. Finally, Breton et al. (2014) showed how health care managers invested in conceptualisation and sensemaking work to modify practices and had the power to mobilise resources to encourage collaborative work during the implementation of integrated local health networks for diabetes in Canada.
While these studies highlight the critical role of entrepreneurial healthcare managers in enhancing partnership logic, there is still lack of strong evidence on successful practices to enhance inter-organisational partnerships in integrated care. Furthermore, integrating care for older adults faces unique implementation challenges (B eland and Hollander, 2011;Threapleton et al., 2017). However, previous studies did not focus on exploring effective practices to enhance the partnership logic in integrated care models for older adults. To address this literature gap, this paper explored a specific research question: what are the main practices enacted by healthcare managers to enhance a partnership logic during the implementation of integrated care models for older adults?

Policy context
Integrated care for older adults is a healthcare policy priority in Canada (MacAdam, 2011). While all Canadian provinces have Beveridgian tax-funded healthcare systems, there are variations in provincial integrated care policies and practices that shape the nature and extent of inter-organisational partnerships. Major policy strategies provide an indicator on when regions are embarking on shifts with regard to inter-organisational partnerships. Two such examples of this kind of policy strategy come from Ontario and Quebec around the same time.
The government of Ontario passed the Ageing at Home Strategy in 2006 to encourage local communities to develop efficient initiatives tailored to the needs of community-dwelling older adults (Kuluski et al., 2016;Peckham et al., 2018). Through this policy, the government of Ontario promoted an organic or bottom-up approach to partnerships by supporting local health and social care provider organisations to work together on a voluntary basis to address common problems. In this context, administrative staff of key provider organisations like Family Health Teams or Community Care Access Centres developed strategies to enhance inter-organisational collaborations and coordination in the delivery of services for their target populations (Breton et al., 2017).
The government of Quebec enacted the Act Respecting Local Health and Social Services Network Development Agencies of 2004 whose main goal was to create 95 Health and Social Service Centres, through administrative mergers of acute care hospitals, long-term care facilities and Local Community Health Centres that mainly provided nursing and social care to community dwelling older adults (Bourque and Quesnel-Vall ee, 2014;Wankah et al., 2018). Furthermore, the government of Quebec explicitly mandated Health and Social Services Centres to lead in the implementation of concertation tables for older adults as a key component of their integrated care model (Poirier et al., 2013;Wankah et al., 2018). Concertation tables were formal structures that aimed at bringing together representatives from Health and Social Services Centres and all relevant provider organisations of their territories around the needs of older adults of their local communities (MacAdam, 2015).

Study design and case studies
This paper offers a supplementary analysis (Heaton, 2008) of a wider international projectthe Implementing Integrated Care for Older Adults with Complex Health Needs (iCOACH) (Wodchis et al., 2018). Specifically, we used a qualitative multiple case study design (Yin, 2018) to explore the process of implementing partnership logic or ideas in integrated care models for older adults in Ontario and Quebec (Breton et al., 2017).
The Ontario site consisted of an integrated care model for older adults that were located in an urban area with a multiethnic and diverse population. Voluntary inter-organisational partnership initiatives emerged between a Family Health Team, an acute care hospital and a Community Care Access Centre. The Family Health Team was a privately owned and funded physician-led multidisciplinary team (family physicians, nurses, nurse practitioners, physician assistant, dietician etc.) that provided a comprehensive range of communitybased primary care services to their local populations. The public funded hospital provided acute care services to local populations of their catchment area. The Community Care Access Centre was a public funded organisation that helped community-dwelling older adults obtain health care and social services in their home by coordinating activities with about 15 community organisations of the territory.
The Quebec site consisted of an urban integrated care model for older adults that were located in an urban area with a multiethnic and diverse population. Government mandated the Health and Social Service Centre to create local partnerships with Family Medicine Groups and community organisations of their territory. The Health and Social Service Centre was a public funded organisation that included a local community health centre and an acute care hospital. The home care unit of the local community health centre multidisciplinary teams of providers (nurses, social workers, occupation therapists etc.) that provided a comprehensive range of community-based services to their population. The hospital provided acute care services to local populations of their catchment area. The Family Medicine Group is a privately owned and funded physician-led multidisciplinary team (family physicians, nurses, nurse practitioners, social workers and pharmacists) that provides primary care services to local populations. Several community organisations provided services such as house cleaning, transportation or meals on wheels to homebound clients.
These two case studies offer unique insights to practices of implementing partnership working for two reasons. First, these models involve collaborations between multiple health and social care organisations, which offers the opportunity to explore organisational change efforts to enhance partnerships across sector boundaries. Second, healthcare managers Implementing integrated care for older adults played active entrepreneurial roles in advancing the transformation of inter-organisational collaborations for these case sites. By analysing the institutional entrepreneurship of healthcare managers in these cases, we may identify unique practices that were enacted to enhance inter-organisational partnerships.

Data collection and study population
We conducted 65 semi-structured interviews of three groups of key stakeholder'spolicy makers, managers and providers (see Table 1)between January 2015 and December 2017. We explored their experiences on efforts to implement partnership working in their respective integrated care models for older adults (Morse, 2015). Policymakers were system leaders at the ministerial level and senior executives of healthcare organisations who had indepth knowledge of strategic and policy issues related to the implementation of integrated care for older adults for the cases under study. Managers were middle level administrative staff of healthcare organisations that were involved in daily tactical management of partner organisations involved in the continuum of care for older adults. Providers were health and social care professionals including physicians, nurses, occupational therapists, or community organisers involved in direct care delivery to older adults living with complex needs. Face-toface interviews were scheduled after each research participant signed an informed consent form. All interviews were audio recorded and transcribed. Each interview ranged from 50 to 90 min.
We also reviewed key policy documents that were relevant to understanding efforts to enhance inter-organisational partnerships in both cases from July 2019 to March 2020. These policy documents included program guides, ministerial action plans or legislations in both cases.

Data analysis
We used a data analysis software -NVivo 11 (QSR International, 2020) -to support data analysis. The institutional entrepreneur theory (Battilana et al., 2009;Garud et al., 2007) provided a theoretical perspective to examine the activities of healthcare managers involved in the implementation of inter-organisational partnerships. We carried out a thematic analysis (Braun and Clarke, 2006) that consisted of coding and categorising key practices that emerged in relation to the implementation of inter-organisational partnerships. A practice was considered as a theme if it fulfilled two criteria: 1) research participants considered the sets of actions as facilitating, improving or enhancing the way organisations worked together in the continuum of care for older adults and 2) research participants perceived that a considerable amount of effort, time or resources was required by healthcare managers to enact the practice. Negative practices that were enacted but hindered collaborations were not included because they were not the focus of this study. Themes were discussed with the research team until we reached a consensus. The final analysis is presented as a narrative synthesis in the findings section.

Findings
In this section we present our research participants' views of various practices that were enacted by healthcare managers to enhance inter-organisational partnerships in their respective integrated care models for older adults. In the Ontario case, by creating a Virtual Ward program, Family Health Team managers introduced the role of physician assistant, negotiated data sharing agreements and co-located staff to enhance inter-organisational partnerships with their local hospital and Community Care Access Centre. In the Quebec case, while struggling to establish a concertation table for older adults, Health and Social Services Centre managers created new roles for community organisers, negotiated service agreements, co-located staff and leveraged joint training to enhance inter-organisational partnerships with community organisations and Family Medicine Groups of their territory.

Ontario
A change in Family Health Team leadership served as a key contextual enabler that triggered the introduction of new partnership ideas as explained by a provider: Three main practices were employed to create the Virtual Ward and to connect it to the hospital: 1) creating a new role for a physician assistant, 2) securing funding for the Virtual Ward and 3) negotiating data sharing agreements across organisational boundaries. After consulting their staff on how best to develop the Virtual Ward, Family Health Team managers soon realised that family physicians were reluctant to participate in a new program that would increase their workloads. Hence Family Health Team managers negotiated with physicians and came to a compromise that consisted of hiring a physician assistant as the main coordinator of the Virtual Ward as explained by a manager: Recognizing though, very quickly, that we could not put this burden necessarily on physicians who are already carrying 1,200 patient rosters; we worked with the Ministry. There was a time, back in Implementing integrated care for older adults 2010, when we could apply for a Health Force Ontario grant to get a physician assistant. The physician assistant was primarily hired to manage patients with complex needs starting with the Virtual Ward. (Manager-ON 05) However, the Family Health Team did not have sufficient budget to hire the physician assistant for the Virtual Ward. As institutional entrepreneurs, Family Health Team managers had the social capital and capacity to mobilise resources to create the Virtual Ward. They did this by reaching out to the Ministry of Health to support them to secure a Health Force Ontario grant that ensured an initial 2-year funding for the physician assistant.
The physician assistant was a key resource to enhance partnership working between the Family Health Team and the hospital. Specifically, the physician assistant participated in discharge planning of complex clients in the hospital. This role allowed the physician assistant to build trusting relationships with clients and the hospital team. Furthermore, the physician assistant worked in close collaboration with Family Health Team nurse practitioners to carry out home visits of clients. However, the physician assistant needed access to client data for effective collaborations with the hospital: I use the hospital's charting, Power Chart, because I am taking Virtual Ward patients from them. So, I need to know the patient's history. And I also need these charts to work with the [X] hospital effectively (Provider-ON 07) Family Health Team and hospital managers enacted two main practices to facilitate the work of the physician assistant. They signed data sharing agreements to allow the physician assistant to consult patients' hospital charts and an electronic notification system was developed to instantly notify the Family Health Team when a Virtual Ward patient was admitted.
At this stage, the Virtual Ward mainly ensured continuous medical and nursing care between the hospital to primary care. However, Family Health Team managers recognised the need to enhance collaborations with organisations delivering social care and community services.
Connecting the Family Health Team to a Community Care Access Centre Since 2008, the Community Care Access Centres has been running a separate programthe Integrated Client Care Programfor community dwelling older adults. The Integrated Client Care Program consisted of Community Care Access Centre coordinators, assessing the social needs of homebound clients and connecting them to relevant community organisations. A critical step towards improving integrated care was the creation of a strategic partnership between the Family Health Team and their local Community Care Access Centre in 2011, as explained by a manager: I would say that one of the foundational pieces that kicked off integration was the partnership between CCAC and [X]FHT. Since that time, we have taken this core, which was integrating CCAC into the Virtual Ward and then supporting home based primary care clients  Two main practices to enhance connectivity between the Family Health Team and the Community Care Access Centre were 1) data sharing agreements and 2) co-location of staff in the Family Health Team.
Difficulties in ensuring adequate client information exchange was an important challenge to connecting the Family Health Team to the Community Care Access Centre, as explained by a manager: When the Virtual Ward was first started, I don't know how many meetings I sat with managers from CCAC. How do we communicate? How does one computer system communicate with the other to help provide care for patients? There were privacy issues. You can't access this computer system. There are two different computer systems. It can't work. And then finally we invited CCAC Care Co-ordinators to come to our [Virtual Ward] rounds and they have access to the system."  Specifically, concerns on the privacy and confidentiality of client data hindered sharing of client information between both organisations. Furthermore, both organisations used different computer systems that further hindered data sharing.
As institutional entrepreneurs, Family Health Team managers worked around these obstacles in two ways. First, data sharing agreements were negotiated with the Community Care Access Centre to facilitate the circulation of client data. These data sharing agreements specified relevant clinically relevant data that could be shared between both organisations, while protecting the privacy of Family Health Team clients. Second, Family Health Team managers agreed to introduce Community Care Access Centre care coordinators for two days a week in the Virtual Ward, where they had access to the Family Health Team computer system and also continued the Integrated Client Care program activities. By co-locating providers and data sharing agreements, the physician assistant worked in close collaboration with Community Care Access Centre care coordinators to ensure continuous medical, nursing, social and community services to homebound clients.

Quebec
The majority of the 37 participants we interviewed explicitly recognised that the 2004 healthcare reforms in Quebec served as a key contextual enabler driving the shift towards a partnership logic as explained by a policymaker: The 2004 reform brought a different view on integrated care. Health and Social Services Centres had to lead in the creation of concertation tables. I think that it facilitated the development of better integration. Better concertation between different partners who needed to coordinate and ensure better fluidity and continuity of services.  Specifically, Health and Social Service Centre managers were required to assume an institutional entrepreneur role by actively establishing concertation tables that brought together public, private and community organisations around the needs of older adults of their local territories. The healthcare managers of the Health and Social Services Centre had the mandate to lead the establishment of a local health network for older adults in partnership with the various organisations and providers located on their local territories. However, our data revealed that Health and Social Services Centre managers faced important challenges in establishing concertation tables and employed various practices to enhance inter-organisational partnerships. These consisted of practices to enhance connections with 1) community organisations and 2) family physicians working in primary care clinics in the community.

Connecting with community organisations
Most respondents were concerned by the limited participation of local community organisations and social enterprises in concertation tables for older adults across Quebec, as explained by a policymaker: I think that managers did not [adequately] understand concertation tables. For example, [Health and Social Services Centre] managers said that they created a concertation table of partners between the hospital and community health centres. Community organisations were not included, social economy enterprises were not included, private nursing homes were not included  Specifically, the respondent pointed out that there was great variability in the way interorganisational concertation tables for older adults were implemented across Quebec. Some Implementing integrated care for older adults concertation tables included community organisations as equal partners, others consulted the opinions of community organisations and other health areaslike the case we studieddid not include community organisations at all. This means that the way Health and Social Services Centre managers interpreted or made sense of policy requirements for concertation tables varied across the territory.
Although Health and Social Services Centre managers faced considerable challenges in engaging community organisations in concertation tables, they deployed three main practices to enhance partnerships: 1) hiring community organisers, 2) negotiating service agreements and 3) organising biannual open days.
In 2005, the Health and Social Services Centre included in this study hired six community organisers to facilitate collaborations with community organisations of their territory. These newly hired community organisers participated in administrative and clinical meetings of the home care unit. Hence, they served as a linkage between Health and Social Services Centre administration and community organisations of the territory. This newly created role gave them leverage to support other community organisations as described by a provider: In fact, I am paid by the Health and Social Services Centre, but I work with community organisations of the territory. I support them in organising their structure, problems of governance, or participation in local projects  Furthermore, Health and Social Services Centre managers negotiated service agreements with local community organisations. Under these service agreements, home care providers could refer vulnerable clients to community organisations for services. However, these were mostly bilateral agreements between the Health and Social Services Centre as the lead organisation and individual community organisations.
Finally, Health and Social Services Centre managers organised biannual open days with community organisations as explained by a manager: We often meet with community organisations. Kind of like open days, where they present their services so that our homecare providers know these organisations and then can get in touch with them. (Manager-QC 01).
The introduction of these biannual meetings allowed community organisations to showcase their products and services to homecare providers. This initiative aimed at fostering collaborations between these organisations in delivering care to older adults of their territory.
Engaging family physicians working in the community Most respondents were concerned by the resistance of family physicians to participate in concertation tables for older adults. The lack of engagement was explained by a policymaker: The efforts we made to make physicians part of the integrated care project. Then it didn't work. And despite everything, when we went to check their opinion on the integrated care model, it was very favorable. [. . .]. Yet, they were not willing to participate in concertation tables.  Specifically, Family Medicine Groups of the territory did not show great interest in participating in concertation tables. Other respondents pointed out that there were no real incentives for Family Medicine Groups to participate in concertation tables.
Despite challenges to include family physicians in concertation tables, Health and Social Services Centre managers enhanced partnerships by 1) co-location of social workers in family medicine groups and 2) joint training initiatives.
In 2016, Health and Social Services Centre managers readily embraced a new government policy that promoted the co-location of social workers in Family Medicine Groups. In fact, social workers were a new type of providers in the interdisciplinary model of family medicine groups. However, some respondents raised concerns about the extent to which social workers could carry out their duties in Family Medicine Groups as explained by a policymaker: It is not entirely clear what the social worker will do in Family Medicine Groups. And even the reception of social workers in Family Medicine Groups is not very enthusiastic for many Family Medicine Groups (Q-Policymaker 005) Specifically, although this policy initiative aimed to improve collaborations with Family Medicine Groups, there were two main issues. There was a lack of clarity of the specific role of public funded social workers in privately owned Family Medicine Groups. Furthermore, initial observations revealed that family physicians were not very enthusiastic to integrate social workers in their Family Medicine Groups.
As institutional entrepreneurs, Health and Social Services Centre managers had the capacity to leverage existing programs to improve connectivity with family physicians. Specifically, Health and Social Services Centre managers promoted joint training of home care providers and family physicians in the context of two programsthe Alzheimer's project and the Behavioural and Psychological Symptoms of Dementia program. The idea was that by promoting interactions between providers from different organisations through joint training programs, this would improve connectivity and build relationships that would improve collaborations between these provider organisations.

Discussion
This paper has explored the implementation of a partnership logic in integrated care models for older adults using the analytical lens of the institutional entrepreneur theory (Battilana et al., 2009;Garud et al., 2007). Acting as institutional entrepreneurs, healthcare managers in Ontario and Quebec sought to transform their organisations by introducing practices that aimed at enhancing collaborations between various organisations in the continuum of care for older adults. Our findings show three similar practices in both cases: 1) introducing new roles (physician assistants in Ontario and hiring community organisers in Quebec), 2) negotiating mutual agreements (data sharing and services agreements) and 3) co-location of staff (introducing care coordinators in Family Health Teams in Ontario and introducing social workers into Family Medicine Groups in Quebec). Other practices to enhance partnerships in these cases include securing new funding in Ontario and biannual open days and joint training initiative in Quebec. We discuss the implications of these findings for institutional change.
A growing body of literature suggests that management may be the Achilles' heel of integrated care (Miller and Stein, 2020). These studies argue that several barriers to the successful implementation of integrated care such as lack of commitment of strategic and operational managers or lack of adequate competencies and skills to manage across organisations are related to managerial roles and functions (Auschra, 2018;Miller and Stein, 2020;Nies, 2006). Our study contributes to this literature by highlighting practices enacted by healthcare managers that assume institutional entrepreneur roles in their efforts to transform their organisations towards more collaborative practices. Our findings show that entrepreneurial managers creatively adjusted the operations of their organisations by introducing new roles, promoting the co-location of staff or negotiating mutual agreements to enhance partnership working. While these organisational strategies are frequently reported in the literature (Auschra, 2018;Nies, 2006;Powell et al., 2015), our findings also showed that entrepreneurial actions of managers were critical to overcoming or working around specific challenges in order to support partnerships. For example, healthcare managers creatively used data sharing or service agreements to improve partnership working within the structural constraints of their respective institutional environments. Implementing integrated care for older adults We identify two main practical implications of these findings. First, institutional change such as enhancing partnership logic in integrated care models for older adults need the participation of managers, acting as institutional entrepreneurs, to support and guide such complex change initiatives. Entrepreneurial managers have the capacity to assess their work environment, introduce new practices or adjust existing practices at clinical, organisational and system levels of their respective healthcare systems (Breton et al., 2014;Forbes, 2012). Healthcare organisations should invest in developing entrepreneurial capacities of employees so as to enhance skills and competencies that facilitate collaboration across organisational and sectoral boundaries (Miller and Stein, 2020;Shaw et al., 2017;Shortell and Kaluzny, 2006). This means that healthcare organisations may either develop proactive recruitment strategies that target people with advanced leadership qualities in key sectors or provide leadership training to existing staff.
Second, our findings show that healthcare managers used multiple and diverse practices in their efforts to enhance partnership logic in integrated care for older adults. These findings suggest that successful institutional change may need integrated implementation strategies that address different areas of the continuum of care for older adults (Powell et al., 2015). For instance, while healthcare policies often focus on joint outcomes as a strategy to engage diverse stakeholders in partnership initiatives (Cook et al., 2007), our Quebec findings show that additional support is required to guide the way healthcare managers interpret and enact policy requirements for partnerships. Furthermore, healthcare managers in Ontario required implementation strategies that addressed regulatory and bureaucratic constraints to information exchange, as well as mobilising resources to establish partnerships. Future research and practice should focus on developing bundles of implementation interventions that address identified barriers to successful implementation (Bauer et al., 2015).
While this study has the merit of reporting firsthand experiences of social actors and their actions to enhance collaborative partnerships in their respective contexts, there are limits to this study. We acknowledge that these findings do not represent all practices enacted by healthcare managers in their efforts to enhance partnership logic; rather they represent salient practices in our cases. Hence these findings can only be generalised to similar contexts. Other literature reviews (Powell et al., 2015) can provide additional information on various strategies to enhance the implementation of healthcare innovations in various contexts. We also recognise the need to further explore barriers and challenges faced by entrepreneurial healthcare managers during the implementation of integrated care models for older adults.

Conclusion
This multiple case study examined effective practices associated with implementing a partnership institutional logic in integrated care models for older adults in two distinct policy contexts. Our findings illustrated that healthcare managers acting as institutional entrepreneurs in their efforts to establish and enhance collaborative practices created new roles, co-located staff, signed mutual agreements and promoted joint training initiatives. These findings also underscore the critical role of healthcare managers in developing strategies to overcome or work around context specific challenges to advance the implementation of integrated care.