Characteristics of the included studies
Of the 13 studies included in this review, seven were based in the USA, three in Europe, two in the UK, and one in Canada. Publication years ranged from 2001–2021 and included 11 qualitative studies, one quantitative study and one mixed method study. The focus of each study explored stakeholder perspectives in cross-sector integrated care and collaboration through population health programs, in community settings. Study topics included: teen pregnancy, domestic violence, breast and prostate cancer, obesity (n = 2), chronic disease prevention, transportation and health, early childhood intervention, social prescribing, and care-dependent seniors. All 13 studies were involved in the integration of health, social care and wider community service and utilized a horizontal integrated care approach. Two studies included the hospital as a partner, the rest included a mix of public/private health and community sectors, and regional/municipal governments. Regarding stakeholder perspectives of cross-sector integrated care initiatives, six studies detailed positive stakeholder perspectives, two studies detailed moderate stakeholder perspectives and five studies detailed negative stakeholder perspectives. Moreover, all studies addressed micro, meso and macro level constructs of integrated care, reflecting the complex needs of the populations targeted within the studies.
Integration Processes
Clinical integration (micro-level)
This dimension represents the heart of integration efforts and is the interface where patients and their families receive care from the health system (Breton et al., 2019). All studies described a clinical integration process to address population health needs. Community health workers (CHW) were utilized in four studies as a key group to mobilize outreach efforts (Enard et al., 2021; Schulz et al., 2001; Shan et al., 2012; Skivington et al., 2018). In addition to providing direct clinical integration, CHWs acted as case managers (Skivington et al., 2018) and were hired to create linkages among organizations (Enard et al., 2021; Skivington et al., 2018). Despite three studies undertaking a community based participatory research (CBPR) focus (Crooks et al., 2018; Enard et al., 2021; Schulz et al., 2001), the use of client participation in co-design was used minimally in studies with and without a CBPR focus (Cicognani et al., 2018; Purcell & Kearns, 2012). Three studies addressed continuity of care as an objective of fluid care delivery. Two of these studies aim to do this through CHW’s (Enard et al., 2021; Skivington et al., 2018), and through cross-sector collaboration (Connaughton-Espino & Reese, 2021). Centrality of client needs was a notable guiding feature in six studies, that sought to integrate and scale client care at micro, meso and macro levels, which was particularly successful when diversified across organizations (Crooks et al., 2018; Lee et al., 2021; Provan et al., 2003; Shan et al., 2012; Van Dijk et al., 2016; Zhu et al., 2019).
Professional integration (meso-level)
Innate characteristics of professionals were a driving force behind many cross-sector integrated care efforts. In some ways, interpersonal characteristics were seen as a supportive element, an added advantage to a well-functioning partnership and influenced acquired learning in partner organizations (Crooks et al, 2018; Shan et al., 2012). Herein, a champion or change agent was identified through skilled leaders in intersectoral collaboration (Skivington et al., 2018; Van Dijk et al., 2016). Yet, the drawback of this emergent leadership was situated in a lack of shared governance or decision making (Zhu et al., 2019). As a result, reliance on interpersonal characteristics to build individual relationships when organizational relationships lacked was common finding (Skivington et al., 2018). This was demonstrated through dependence on CHW’s to generate community awareness and trust. The skills of CHW’s can be amplified as sole ‘problem solvers and emotional supports’, while the responsibilities of others in the partnership are negated (Purcell & Kearns, 2012; Schultz et al., 2001).
Shared vision between professionals was often established from the outset in collaborative partnerships (Cicognani et al., 2018; Connaughton-Espino & Reese, 2021; Crooks et al., 2018; Middleton et al., 2013; Shan et al., 2012; Schulz et al., 2001). These early formalities can be captured through agreements of interdisciplinary collaboration or memorandums of understanding (Cicognani et al., 2018; Connaughton-Espino & Reese, 2021). However, early establishment of shared vision did not necessarily equate with collaborative success in short- or long-term project indicators (Connaughton-Espino & Reese, 2021; Middleton et al., 2013; Schulz et al., 2013). Of particular importance is who is creating a shared vision from the outset, where excluding key community organizations led to detrimental outcomes (Schulz et al., 2013). Shared vision was found to expand and deepen over time in positive, collaborative working relationships (Cicognani et al., 2018; Crooks et al., 2018). Furthermore, interprofessional education provided an avenue to facilitate increased awareness of services, inner workings of health systems, and cultural relevance if all parties were present (Enard et al., 2021; Purcell & Kearns, 2012; Shan et al., 2012). When interprofessional education narrowly addressed access to services, rather than cultural community-based teachings, project scope and reach were limited (Enard et al., 2021, Purcell & Kearns, 2021). Similarly, multidisciplinary guidelines provided a neutral reference for organizations to situate their work, advance understanding in evidenced based practice, strengthen relationships, and increase community capacity (Enard et al., 2021; Lee et al., 2020).
Organizational integration (meso-level)
Nomenclature used to describe cross-sector integrated care initiatives varied widely across studies. The term ‘collaboration or collaborative’ was used most frequently (Crooks et al., 2018; Provan et al., 2003; Van Dijk et al., 2016; Zhu et al., 2019), followed by ‘partnership’ (Cicognani et al., 2018; Schulz et al., 2001; Shan et al., 2012), then ‘program’ (Middleton et al., 2013; Skivington et al., 2018), ‘initiative’ (Connaughton-Espino & Reese, 2021) and ‘coalition’ (Purcell & Kearns, 2003). Terms were defined and labelled according to overarching goals of the integrated group or partners themselves. For example, Crooks et al., (2018) referred to their group as a collaborative, a small agile group of individuals, dedicated and responsive to one another needs. Shan et al., (2012) equally described accountability measures taken in community partner selection, however partnership inclusivity remained at the forefront. Grounded in anticolonialism and indigenous health principles, Shan et al., (2012) chose to refrain from using the term ‘collaborative’. Even among partners within the same study various terms were used dependent on perceived goals and desires. One hospital partner referred to their interaction with community services as a collaborative, when further analysis by Zhu et al. (2019) revealed the level of engagement was in fact cross-sector interaction, a unidirectional level of outreach. In addition, poor interorganizational governance between the health and community sectors was causal for group reorientation and labelling. Purcell & Kearns (2012) demonstrated how the community sector can break ties with an organizational working group and refer to themselves as a ‘coalition’ in effort to build their own capacity.
Attention, dedication, and alignment of interorganizational governance played a significant role in the positive perspectives of stakeholders in community centered integrated care. The opportunity for all partners to meet regularly, co-design programs and share decision making allowed for greater understanding of capabilities within the group (Cocgnani et al., 2018; Crooks et al., 2018; Middleton et al., 2013; Shan et al., 2012). Moreover, attention placed on the general SDOH within a community, rather than a narrow programmatic focus allowed stakeholders to unite through a common vision, act with honesty and value local, bottom-up experiences of health promotion (Cocgnani et al., 2018; Crooks et al., 2018; Shan et al., 2012). Progress reports and annual work plans provided added support and direction within partnerships to create learning organizations to advance collective research and program development (Crooks et al., 2018; Lee et al., 2021). Conversely, the exclusion of key working group partners from the outset fractured trust and momentum in advancing project goals (Purcell & Kearns, 2012; Shultz et al., 2001). Negative perspectives of stakeholders were found in studies suggested to be based in CBPR and consensus building but lacked attention to these efforts throughout (Connaughton-Espino & Reese, 2021; Schulz et al, 2001). Negative perspectives among stakeholders were also found in studies whose outcome presented as value creation for organization. Herein, a beneficial outcome for the organization is sought after verses overarching population health features. This demonstrated paralysation of community social capital and health equity, as well as disabling trust, at the hand of academic, health and municipal partners (Enard et al., 2021; Purcell & Kearns, 2012; Schulz et al., 2001).
System Integration (macro-level)
Equal to the importance of robust interorganizational governance in community centred integrated care, is diverse stakeholder management. Programs, partnerships, and collaborations that upheld commitment from diverse stakeholders were resoundingly positive through stakeholder perspectives. These initiatives included engagement from municipalities (Cocgnani et al., 2018; Crooks et al., 2018; Shan et al., 2012), and public and private organizations (Lee et al, 2021; Provan et al., 2003; Zhu et al., 2019). Importantly, each study referenced above referred to community organization representatives as key stakeholders. The involvement of distant funding partners and government legislators did not necessarily equate to program longevity or positive experiences among project partners. This was particularly true when direct funding partners and those in decision making authority were not involved in proximal stakeholder management (Connaughton-Espino, 2021; Enard et al., 2021; Shultz et al., 2001; van Dijk et al, 2016). Funding that was diversified and secured created opportunity for program partners to plan longitudinally, develop trust and cultivate relationships with each other and their communities (Cocgnani et al., 2018; Crooks et al., 2018; Lee et al, 2021; Provan et al., 2003; Zhu et al., 2019).
Programmatic Alignment And Leadership Style
The context in which programs are created directly sets them up for success or failure. Those designed to respond to gaps in care were often seen as bridging rather than building sustainable solutions. This is detailed through reliance on CHW’s in the context of deepening austerity measures (Skivington et al., 2018), and competitive tender practices to appoint new partners in care, which limits foundational integrated care development (van Dijk et al., 2016). Programs that were created through existing structures and systems, governed by institutional polices, had a greater reach and opportunity to evolve through stakeholder acceptance and comfort (Crooks et al., 2018, Lee et al., 2021; Shan et al., 2012). Although diverse stakeholder representation was deemed beneficial at the proximal level in community centred integrated care, Middleton et al., (2013) calls for a single leader to champion efforts. In this example, public health and local authorities worked ‘side by side’, while the community voice was absent in leadership representation. Conversely, Crooks et al., (2018) called for distributed leadership, that is emergent and context dependent. Reliance on trust and foundational relationships was seen as pivotal to enacting this leadership style, which requires shared project completion by partners in varying roles, also known as socially distributed cognition (Crooks et al., 2018).
Functional Integration (macro-level)
Indices of functional integration were found in 9/13 studies. A variety of human resource management structures were found, where community partners or representatives were included in joint planning and hiring processes (Cocgnani et al., 2018; Crooks et al., 2018; Lee et al., 2021, Shan et al., 2012; van Dijk et al., 2016). Those that employed community organizations as a third party to complete outreach and engagement work had less favourable outcomes, and uniquely all stakeholders in these arrangements spoke to this power imbalance as projects progressed (Connaughton-Espino & Reese, 2021; Enard et al., 2021; Purcell & Kearns, 2012; Shultz et al., 2001). Although few projects demonstrated regular feedback and performance indicators (Crooks et al., 2018; Lee et al., 2021), recognition of their importance to conduct quality improvement and celebrate success was noted (Middleton et al., 2013; Zhu et al., 2019). Projects that had regular feedback mechanisms reported positive perspectives from stakeholders (Crooks et al., 2018; Lee et al., 2021), whereas the development of joint evaluative frameworks, with community partners, to address soft versus hard performance indicators was seen as a necessary endeavor to create neutral ground among cross-sector partnerships (Middleton et al., 2013; Zhu et al., 2019). Despite the complex and tailored approaches that integrated care requires, several studies called for a framework to guide professionals toward developing mutually agreed upon goals to sustain relationships in urban and rural cross-sector partnerships (Enard et al., 2021; van Dijk et al., 2016; Zhu et al., 2019).
Normative Integration (macro-level)
Among the studies, there was a direct relationship between the level of community engagement and the resulting perceptions of stakeholders. Specifically, the further along the continuum of community engagement, the more positive were stakeholders’ perceptions of their integrated care experiences (Table 2). Although normative integration was referred to explicitly in only one study (van Dijk et al., 2016), an abundance of shared stakeholder attributes, values and desires presented across the studies. The concept of trust was found in 9/13 studies and was seen as a multilayered construct. Cicognani et al. (2018) found that trust was associated with health partnership performance, and tied to efficacy, commitment, and a sense of belonging within community. Similarly, deepening and building relationships lead to trust, and was present in all levels and systems of care integration (Crooks et al., 2018; Lee et al., 2021; Provan et al., 2003). Among all studies, trust was coveted but it was unachievable in some settings due to power inequities, lack of time or investment within partnerships (Connaughton-Espino & Reese, 2021; Enard et al., 2021; Schultz et al., 2001; Skvington et al., 2018). Once trust was fractured, it was difficult to repair (Connaughton-Espino & Reese, 2021; Enard et al., 2021; Purcell & Kearns, 2012; Schultz et al., 2001; van Dijk et al., 2016). This is a sentinel finding given the importance of creating a collective attitude and shared vision at the beginning of a partnership to support momentum, longevity, and integration into existing partnerships (Crooks et al., 2018; Shan et al., 2012). Several studies reviewed the importance of being flexible, reinventing roles as needed and transcending domain perceptions (Crooks et al., 2018; Lee et al., 2021, Provan et al., 2003; Purcell & Kearns, 2012; van Dijk et al., 2016). Plausibility to transcend one’s professional domain perceptions comes with good governance, partnership security and visionary leadership, however few studies identified such a leader (Crooks et al., 2018; Lee et al., 2021). Lastly, integrated care projects within studies were conducted between two – eight years, two years being the average length of time. There does not appear to be a correlation between project duration, level of community engagement and stakeholder perception. All studies were enacted without a sense of urgency identified.
Table 2
Stakeholder perspectives of cross-sector integrated care in relation to community engagement level
Level of Community Engagement
|
Positive perspectives
|
Moderate perspectives
|
Negative Perspectives
|
|
Cicognani et al.
|
Crooks et al.
|
Lee et al.
|
Provan et al.
|
Shan et al.
|
Zhu et al.
|
Middleton et al.
|
Skivington et al.
|
Connaughton-Espino et al.
|
Enard et al.
|
Purcell & Kearns
|
Schultz et al.
|
Van Dijk et al.
|
Outreach
|
|
|
|
|
|
|
|
|
x
|
x
|
x
|
x
|
|
Consultation
|
|
|
|
|
|
x
|
x
|
x
|
|
|
|
x
|
x
|
Involvement
|
|
|
x
|
x
|
|
x
|
|
|
|
|
|
|
|
Collaboration
|
x
|
|
|
|
x
|
|
|
|
|
|
|
|
|
Shared leadership
|
|
x
|
|
|
|
|
|
|
|
|
|
|
|