Introduction

Integration of healthcare services has been a policy goal in jurisdictions around the world for decades [1, 2]. Specifically, the integration of primary care (PC) into broader health systems and social services has been a focus [3,4,5,6], with PC positioned as a key access and coordination point in a broader move to deliver Primary Health Care (PHC) [7,8,9,10,11]. Efforts to achieve integration have approached it as an end state in which diverse clinicians come together to provide care to those with complex health needs, while eliminating redundancies or gaps in service [12]. From this broad and clinically focused definition, a literature has developed that studies integration as it is attempted at micro-, meso-, and macro-levels [13].

The present paper analyzes the meso-level integration activities of PC participants as they sought to accomplish more by collaborating across the organizational silos of their health system [14, 15]. Meso-level here is distinct from both the clinical operations of the micro-level, and the macro-level efforts of national or international organizations seeking high-level policy changes. We describe the integration efforts of meso-level participants from implementation-oriented departments of the government, operations-focused PC governance organizations, and associations or colleges representing PC providers in Alberta, Canada. We describe them attempting to ‘do’ integration at a time when relationships were tenuous, and at a particular moment in the COVID-19 (C19) pandemic. By showing a particular group of PC participants signaling one another, sharing ways of thinking, and enacting cultural values [16, 17] and so achieving greater continuity of planning and care delivery in the pandemic response [18], our aim is to make a broader contribution to the practice of integration.

While a robust body of literature describes the facilitators of (and barriers to) integration at the micro-level [19,20,21], the effects of local culture on the uptake of integration-focused improvement initiatives remain ‘underspecified,’ [22].Similarly at the meso-level, the evidence available to support actually doing complex culture change [23, 24] has been described as “thin” [25]. This is to say the literature often invokes high-level concepts like “culture”, “relationships”, or “communication” as important factors, but rarely examines the ways in which these concepts are enacted, particularly as facilitators between institutions or organizations.

In the analysis that follows, we illustrate how participants practically enacted culture, relationships, and communications in a way that fostered integration. Our work here aligns with that of PC scholars who have sought to understand integration as an emerging set of practices [18]. Specific mechanisms of integration, such as quality improvement measures [26] performance feedback [27], incentives and financing models [13, 27,28,29,30] and governance structures to promote accountability [5, 13, 26, 31] have been identified. Across these mechanisms, and the micro-, meso-, and macro-levels of intervention, the literature suggests the importance of establishing a pro-integration culture [32] that emphasizes inter-organizational relationships [13, 26], and facilitates open communication [27, 33]. Beyond these high-level encouragements to pursue underspecified goals, the evidence for how to go about this culture, relationship, and communication work is thin. What follows draws pragmatic lessons in ‘doing’ meso-level integration from the experiences of Alberta’s PC participants.

Background

PC in Alberta is directly financed by the province’s ‘single payer’ [34, 35] Ministry of Health (MoH), and most care is delivered by independent family physicians who bill the government on a fee for service (FFS) basis. The fees a family physician may charge are negotiated between the Alberta Medical Association (AMA) and the MoH. Over the last decades, remuneration, the viability of PC practices as businesses paying their own overhead, and efforts to ensure access to PC for increasingly complex patients in a range of settings across the province, have made those negotiations contentious, while also leading to significant innovation [33]. We describe this innovation in more detail below.

Alongside the independent FFS model of PC delivery, the province operates the largest centralized healthcare system in Canada, with over 650 facilities managed by a single health authority: Alberta Health Services (AHS). AHS is charged by the MoH with delivering care in five geographically-based ‘health zones,’ with facilities providing acute, long-term, and urgent care. A small PHC-focused unit inside AHS is devoted to achieving PC integration as a policy objective [36]. In addition, innovative bridging organizations called Primary Care Networks (PCNs), financed through capitation, act as key points of contact between central AHS and independent family physicians who opt-in to PCN membership [33].

As C19 arrived in Alberta in March of 2020, relations between the MoH and the AMA were cold. The negotiation of virtual care billing codes to facilitate community management of C19 patients took place at a moment of deep chill in the relationship between independent PC and the central system [37,38,39]. Nonetheless, over time, the relationship warmed and mutually acceptable arrangements were made [5]. The data we present below describe activity at a subsequent moment in what we argue is an ongoing, and perhaps universal, cycle of freezing and thawing relationships in the integration context.

Methods

An ‘interpretive description’ approach [40] was taken to analyse interviews with key informant participants. Interpretive description focuses on identifying applied knowledge and allows for specific attention to participants’ institutional commitments and perspectives [41, 42]. It provides insights not just into areas of commonality but also areas of disagreement among participants, with an eye on unearthing pragmatic suggestions to improve policies and outcomes [40, 43]. With research showing that purposive, rather than random, sampling is effective at drawing information from small samples [44, 45] focused interviews with a limited number of key informants were conducted. This approach has been validated as increasing the likelihood of attaining data saturation [46], with evidence indicating that when a qualitative study: is more specific in its aims; targets informants with specific knowledge; and relies on interviews of sufficient length and quality, saturation can be achieved with relatively low sampling [47].

Semi-structured qualitative interviews with key informants were conducted as part of a broader project seeking to understand the experiences of healthcare participants from across the provincial health system as they responded to C19. Participants in the broader project were initially recruited purposively based on their administrative or clinical roles and involvement in the creation or implementation of pandemic response policies. A snowball sampling strategy was adopted thereafter as we identified areas of policy innovation and implementation activity. Our particular focus was on the challenges encountered as policy was implemented across the boundaries of health system organizations [removed for peer review purposes] [48]. In this way administrative, physician, and patient leaders from a range of micro-, meso-, and macro-level organizations were interviewed in the full sample, with data collection focused on moments of policy transfer or implementation that spanned boundaries or sought to coordinate and integrate activity [removed for peer review purposes] [49, 50].

From that full sample of semi-structured interviews (n = 127) with 114 unique participants, the present analysis focuses on 11 interviews with nine key informants from meso-level organizations conducted by [Removed for Peer Review Purposes] between January and March 2022. The focus of these interviews was on the planning for, and response to, the Omicron wave which began in September of 2021. The Omicron wave was, in this sense, an episode identified in the broader sample of interviews that could be analyzed to identify themes and draw lessons in doing integration at the boundary between PC and the central system. While anonymity concerns preclude us from offering details about the meso-level organizations that these key informants were associated with, Table 1 provides an overview of their alignment with either independent PC or central-system aspects of the integration context.

Table 1 Key Informant Interview Participants in Methods
Full size table

The interviews in the present paper averaged 65 min in length and were digitally recorded and transcribed for analysis. [Removed for peer review purposes] analyzed the data to identify themes. Analysis was supported by MAXQDA 2022 software and deployed an inductive coding approach to render an interpretive description of key informants’ integration experiences and activities at the boundary between PC and the central health system. The authorial team analyzed the data iteratively, expanding, collapsing and merging themes to arrive at the final analysis. We summarize passages from the verbatim transcripts in our analysis and provide access to the full quotes in Boxes at the close of each section in the Results. We attribute the quotes to participant numbers ranging from 18 to 114 (see Table 1). This research obtained ethical approval from the Conjoint Health Research Ethics Board at the University of Calgary (REB20-0371). All participants provided written and verbal consent.

Results

Contextual freeze-thaw cycle

Participants felt the pandemic had catalyzed significant connectivity across the PC environment, increasing interactions between the various parts of the health system (Table 2, Quote 1). However, relations between the MoH and PC were decidedly cool in the summer of 2021. MoH officials were taking direction from a government that all-but declared the pandemic over as it vowed to make the season the ‘best ever’ [50, 51], and PC leaders felt they had been taken advantage of. Just a week before, they had begun work on a plan for delivering PC under endemic conditions, and were caught off guard by government moves to suspend mass testing and return to ‘normal’ (T2, Q2). Beyond surprised, PC leaders felt snubbed, betrayed, and like the engagement and consultation process to integrate PC into the ongoing pandemic response was disingenuous (T2, Q3). Some acted on these feelings by walking away from the conversation entirely (T2, Q4). In a contextual cycle of freezing and thawing relationships between the MoH and PC, then, June 2021 marked a particularly cold point and a low in meso-level PC integration into the pandemic response.

Table 2 Verbatim quotes supporting Freeze-Thaw Cycle
Full size table

Months later, however, the cycle reversed. According to participants, the thaw in relations, and so increase in PC integration, was attributable to a range of exogenous factors: (1) The arrival of the Omicron variant that defied the optimism of the summer; (2) A provincial cabinet shuffle (T2, Q5) introduced a new consultative style (T2, Q6); (3) A new Assistant Deputy Minister at the MoH took up the health portfolio and made good on the new Minister’s consultative style (T2, Q7); and (4) The hiring of a new Deputy Chief Medical Officer of Health (DCMOH) who brought the various participants back to the table, provided reliable information, and lived the consultative ideal (T2, Q8). Under these conditions, as the ‘best summer ever’ became the autumn of Omicron, a freeze-thaw cycle in Alberta’s integration context moved into a warmer period.

Doing pro-integration culture

Achieving a pro-integration culture involved sending value signals through staffing choices; personal expressions of commitment to the values of integration; and ensuring communal commitments to those values. The hiring of a new DCMOH – at the request of an overstretched Chief Medical Officer of Health (Table 3, Quote 1) – was a key moment in the thaw. The Chief sent a signal by creating a position focused on PC integration and hired someone who was able to deliver on her priorities and consultative style (T3, Q2).

Table 3 Verbatim quotes supporting Doing Pro Integration Culture
Full size table

This signal that PC was valued was received as intended in both the PHC-focused unit inside AHS and on the front lines of PC. The AHS-PHC unit saw the choice to hire a family doctor into the DCMOH role as a step towards legitimate, good-faith engagement (T3, Q3), and leaders in the independent PC world felt, with the appointment, that they had an easily-engaged advocate in the halls of power (T3, Q4). Similar staffing choices, and so signals of the value of integration work, were sent by the AHS-PHC unit. These were also well received, creating a sense of team and collective enterprise (T3, Q5). Thus, doing integration involved successfully sending a value signal through staffing and resource choices. The values of pro-integration culture were made real through the creation of specific jobs and a commitment to integration work.

Personal expressions of commitment to integration were similarly seen as enactments of a shared value and common priority. Expanding on the importance of consultation, PC leaders emphasized how words and deeds that showed personal commitments to integration built trust over time (T3, Q6). Specific actions beyond being consultative and collaborative (T3, Q6) included: assuming good faith and intentions (rather than bad-faith and selfish intent) on the part of others (T3, Q7), and indicating a desire to work together by not shutting others down (T3, Q8).

In addition to individual signals of commitment to the collaborative values of pro-integration culture, participants also described communal expressions through shared work. A PC leader used a collective pronoun to describe how the work of the autumn thaw period felt like the various participants were striving for common goals (T3, Q9). Another PC leader invoked collaboration, contrasting it with directive command and control approaches, as the enactment of integration’s cultural values. They also noted that in ‘playing more collaborative’ and ‘being less directive,’ their MoH counterparts were engendering trust (T3, Q10). Trust here was accomplished by the assumption of good faith noted above, and by the inclusion of more voices and perspectives (T3, Q11-2).

Doing integration relationships

Working at the relationships that contributed to, or were able to benefit from, a thawing integration environment required participants to follow through on their rhetoric, and persevere in proving themselves to their counterparts. From the perspective of both MoH and PC leaders, following through on stated intentions was key to doing the relationship work that supported integration. As part of enacting the pro-integration value of collaboration, all those involved needed to do what they had promised (Table 4, Quote 1). PC leaders judged their MoH counterparts’ follow through work against real-world outcomes. Those outcomes might include changed public health messaging that incorporated PC perspectives; or a willingness to revisit the billing codes that kept independent PC financially viable (T4, Q2); or supplying government purchased personal protective equipment to PC practices (T4, Q3). This was not to say PC leaders demanded follow-through on all their issues, but rather that doing the relationship work of integration required an openness to discussion and making good on commitments (T4, Q4). Following through, then, was a key to enacting a healthy integration relationship.

Table 4 Verbatim quotes supporting Doing Integration Relationships
Full size table

Similarly, participants noted that perseverance in the integration relationship was important. As a PC Leader described it, perseverance was a steady ‘grind’ of relationship work that, even if it yielded more success in a thaw period, required effort even during cooler periods (T4, Q5). An AHS-PHC leader described how perseverance bridged the freeze periods, setting the possibility for integration when the thaw came, and previous misgivings could be set aside (T4, Q6). This characteristic of perseverance was described as an ongoing struggle to have PC recognized and valued (T4, Q7) and a heroic undertaking in the face of frozen relationships (T4, Q8). Through perseverance, despite the odds and the summer’s freeze, PC leaders were able to immediately take advantage of the autumn’s thaw. They could take not just ideas, but fully formed plans that had been rejected earlier, and present them to newly attentive eyes (T4, Q9). Being present and seeking constantly to prove the value of PC in the pandemic response were essential to enacting integration. Persevering with relationships that might look dormant, or worse, was of central importance.

Doing integration communication

Participants described specific ways of communicating to achieve integration. These included: practicing bi-directional communication; and actuating a mix of formal and informal communication channels. Practicing bi-directional communication was a key behavior underpinning the thaw in relations that all participants described. As parties on both sides showed they were willing to listen, as well as to talk, the communication pattern was perceived both as more cordial (Table 5, Quote 1), and as having shifted from uni-directional commands from the MoH, to a reciprocal flow of co-designed ideas and feedback from PC on implementation efforts (T5, Q2). The shift was seen as central to achieving two-way consultation, and so the integration of PC into policy formation (T5, Q3).

Table 5 Verbatim quotes supporting Doing Integration Communication
Full size table

A final practical step for achieving communication that supported integration involved accessing a mix of formal and informal channels. While the autumn thaw was characterized by the creation, or resurrection, of formal committees, (T5, Q4) bi-directional consultative communications also occurred in unofficial conversations (T5, Q5). These informal channels took advantage of individual PC leader’s broader social and professional networks, drawing more perspectives and ideas into the relationships that were being cultivated (T5, Q6). Participants emphasized the importance of informal social relationships that extended beyond the formal committee work, describing how those relationships and communication channels were central to achieving integrative goals (T5, Q7).

Discussion

Our data suggest there is an ebb and flow of people and relationships in the meso-level integration context. Following our participants, we have deployed a temperature metaphor to describe this pattern, calling it a ‘Freeze-Thaw’ cycle. At one level this cycle is the product of factors exogenous to the integration effort, like the C19 Omicron wave, or imposed changes in leadership. At another level, and borne out by our data, organizations can do and learn more about integration beyond relying on exogenous shocks like pandemics to shake things up and alter the ‘temperature.’ In Alberta, the pandemic highlighted pragmatic temperature-raising activities that supported meso-level integration.

The relationships between meso-level organizations in the integrated care literature tend to be conceptualized as more or less deep. Greater depth in those relationships is in turn seen as allowing the organizations to pool skills and expertise and so achieve integration and improve care [13, 51, 52]. Our data, described through a temperature metaphor, suggest this depth varies cyclically. Whether integration relationships are freezing and thawing, or ebbing and flowing, further research is required to understand if and how the cycle presents itself across jurisdictions. While we can be confident that publicly funded health systems are influenced by exogenous factors such as elections, leadership turnover, or pandemics [5, 53], the way these factors shape meso-level participants in varying locations as they ‘do’ integration is unclear. For now, and assuming some level of generalizability from our key informants’ experience of Alberta’s Omicron Freeze-Thaw, we turn to providing pragmatic options for those in other jurisdictions to consider.

The literature on integrating PC into health systems emphasizes the importance of participants espousing a pro-integration culture [32] that focuses on strong relationships [13, 26] and open communications [27, 33]. However, practical advice on doing this culture – enacting its relationships and communicating its values – is thin [54]. Culture, generally, can be defined as the negotiation and expression of commonly held values and norms [55]. In the specific case of a ‘pro-integration culture’ those values and norms have been specified [56], and their negotiation and expression involve community members: visibly manifesting them; sharing ways of thinking; and sharing deeper unsubstantiated assumptions about how integration works [16, 17].

Our data suggest that making pro-integration culture real requires sending and receiving signals that a collaborative working environment is valued. These signals can be sent through: staffing and other resource decisions; matching rhetorical commitments with collaborative actions; shifting from uni-directional talking to bi-directional talking and listening; and activating a mix of formal and informal communication channels. These visible manifestations of individual and group commitments to collaboration are the product of more than a mere attitude switch. From walking the talk to engaging in ‘active listening’ [57, 58], these are enactments of shared mental models [59,60,61] and the trust that becomes possible when one assumes others are acting in good faith, and when one relinquishes a measure of control over the conversation by adding voices to it. Beneath the visible signs of commonly held values, then, Alberta’s meso-level participants formed what the literature refers to as ‘moral community,’ holding one another mutually accountable for following through on the rhetoric and ideals of collaborative integration culture [62].

The success of this moral community and the flourishing of this pro-integration culture were founded in trust. Trust is characterized by vulnerability [63]. To trust someone is to be vulnerable to them without proof or assurances that they will do what we hope they will do [64]. In this sense, Alberta’s MoH participants made themselves vulnerable to their PC counterparts when they began assuming good faith and positive intent [65], rather than bad faith and selfish intent as had prevailed during the preceding freeze of the contextual cycle. The doctors might have been acting in bad faith, but the administrators chose to trust their intentions. Similarly, the MoH participants exhibited vulnerability when they increased the size and scope of the consultation table, thereby relinquishing some of their control of the conversation. Embracing these sorts of vulnerability generated trust. As such, embracing vulnerability appears to be the sort of practical action that can be taken to advance integration. Perhaps even more importantly, those seeking to do integration may wish to consider the value of perseverance. Our data suggest that persevering – at sending the right signals, at maintaining relationships, at active listening as well as talking, at fostering formal and informal communications, at embracing a vulnerability that cedes control but gains trust – is central to achieving integration. It is the quality, applied to all these pragmatic behaviours, that appears to support warming and prevent cooling in the integration context.

Conclusions

Achieving PC integration likely involves more than passive resignation to the effects of exogenous and contextual factors. The proactive work of enacting specific cultural, relationship, and communication behaviors also appears to contribute. Those seeking to integrate at any given moment in their own context may consider a range of practical activities that enact culture and engender trust: from sending signals with staffing, to connecting rhetoric with demonstrable action; from showing vulnerability to achieve trust, to persevering at relationships. Enacting these visible behaviors, and so the collaborative values that subtend them, appears to support integration regardless of the external shocks, internal staff turnover, and relative coolness of the local freeze-thaw cycle.