Article Text

Original research
Lessons learnt from the implementation of new models of care delivery through alliance governance in the Southern health region of New Zealand: a qualitative study
  1. Gagan Gurung1,2,
  2. Chrystal Jaye1,3,
  3. Robin Gauld2,4,
  4. Tim Stokes1,2
  1. 1Department of General Practice and Rural Health, Dunedin School of Medicine, University of Otago, Dunedin, New Zealand
  2. 2Centre for Health Systems and Technology (CHeST), University of Otago, Dunedin, New Zealand
  3. 3CARE Research Theme, University of Otago, Dunedin, New Zealand
  4. 4Otago Business School, University of Otago, Dunedin, New Zealand
  1. Correspondence to Professor Tim Stokes; tim.stokes{at}otago.ac.nz

Abstract

Objectives To explore the process of implementation of the primary and community care strategy (new models of care delivery) through alliance governance in the Southern health region of New Zealand (NZ).

Design Qualitative semistructured interviews were undertaken. A framework-guided rapid analysis was conducted, informed by implementation science theory—the Consolidated Framework for Implementation Research.

Setting Southern health region of NZ (Otago and Southland).

Participants Eleven key informants (Alliance Leadership Team members and senior health professionals) who were involved in the development and/or implementation of the strategy.

Results The large number of strategy action plans and interdependencies of activities made implementation of the strategy complex. In the inner setting, communication and relationships between individuals and organisations were identified as an important factor for joint and integrated working. Key elements of a positive implementation climate were not adequately addressed to better align the interests of health providers, and there were multiple competing priorities for the project leaders. A perceived low level of commitment from the leadership of both organisations to joint working and resourcing indicated poor organisational readiness. Gaps in the implementation process included no detailed implementation plan (reflected in poor execution), ambitious targets, the lack of a clear performance measurement framework and an inadequate feedback mechanism.

Conclusions This study identified factors for the successful implementation of the PCSS using an alliancing approach in Southern NZ. A key enabler is the presence of a stable and committed senior leadership team working through high trust relationships and open communication across all partner organisations. With alliances, partnerships and networks increasingly held up as models for integration, this evaluation identifies important lessons for policy makers, managers and services providers both in NZ and internationally.

  • Health policy
  • Organisation of health services
  • Organisational development
  • QUALITATIVE RESEARCH

Data availability statement

No data are available. Full deidentified interview transcripts will not be shared. Informed consent, in line with the approving ethics committee, only allows for the use of deidentified extracts within research reporting and writing to maintain participants’ privacy based in a defined regional area and population, thus making their identification with full transcripts more likely.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

STRENGTHS AND LIMITATIONS OF THIS STUDY

  • This study contributes to a currently small body of research using implementation science theory (Consolidated Framework for Implementation Research, CFIR) to study new models of care delivery using an alliancing governance model.

  • The use of the CFIR helped us to illuminate contextual factors and understand the complex interplay between the context and implementation process of the strategy intervention.

  • The study’s participants had a governance and senior managerial role and were directly involved in developing and/or implementing the primary and community care strategy.

  • The study was conducted in partnership with the local health system, which was helpful in facilitating the sharing of findings and feedback to the Southern health system.

  • Use of a rapid analysis approach was helpful in providing prompt actionable feedback to the local health system, but it might risk missing nuances of data.

Introduction

Health systems worldwide and in New Zealand (NZ) are facing a number of challenges, which are likely to intensify in the future.1 2 A pressing challenge is the need for better integration and coordination of services.1–6 Reducing fragmentation and achieving integration is a key response and, in NZ, a goal of recent health policy and system reforms.1 4 7

An approach used in NZ at the local health system level from 2013 to 2021 to promote the integration of healthcare across primary and secondary care was that of alliancing.8 Alliances bring all key providers within a local health system together in the process of governing healthcare design and delivery, with a focus on building whole-of-system service designs.9 10 Derived from the construction industry, this concept has been used in healthcare in NZ and other countries such as the UK and Australia.11 12 Alliancing promises a high trust, low bureaucracy way of working between organisations.11 13 Ideally, members of an alliance should have the capacity to bring resources to the alliance table so decisions can be implemented, and to put aside sectoral interest to work collaboratively towards a joint goal and take a whole-of-system approach to planning and decision making.9 11 The evaluations of past initiatives developed and implemented through an alliance approach in NZ and via similar initiatives (eg, accountable care organisations in the USA and Vanguard programme in the UK) internationally show some promise in terms of improving integration, although it needs to be noted that the health system context in which such initiatives are being implemented is often both complex and changeable.9 14–16

In the Southern health region of NZ (see box 1) the Southern District Health Board (SDHB) and WellSouth Primary Health Organisation (PHO), working together through a formal contractual alliance (Alliance South), developed a Primary and Community Care Strategy (PCCS) (see box 2) ‘to do things differently in primary and community care.’17 The strategy, launched in 2018, reflected the commitments of the two alliance partners and priorities for improving primary and community care in the Southern region. Priority action areas for the delivery of the strategy were new models of care workstreams (eg, healthcare home, home team, community health hub and locality network) and enabling infrastructures (eg, governance and leadership, workforce capability and culture, funding and contracting). Implementation of the PCCS was carried out by Alliance South with joint governance from leaders in both the SDHB and PHO through the Alliance Leadership Team (ALT). The strategy’s action plan included establishing a culture of continuous improvement supported by the monitoring and rapid evaluation of new initiatives.18

Box 1

NZ Southern health system

During the period of research, the NZ Southern health system comprised a DHB, being the largest geographic region out of 20 DHBs across the country. There was a single PHO in the region. The DHB and PHO served just over 300 000 people with 40% living rurally. The DHB had the overall responsibility for planning and funding in the region and owned public hospitals. The PHO received funding from the DHB to support primary care and affiliated general practices. The region has two main hospitals and six small regional hospitals. In mid-2022, DHBs were disestablished with functions absorbed into a new national body, Health NZ, as part of major health reforms.

DHB, district health board; PHO, primary health organisation; NZ, New Zealand.

Box 2

The PCCS: vision and strategic goals

The primary and community care strategy (PCCS) provides a vision for primary and community care in the Southern health system. It recognises the challenges the health system faces in responding to the changing needs of the community, increasing pressure on the health workforce and the responsibility to provide equitable access to services across the large and diverse district.

The strategy and action plan was developed jointly by SDHB and WellSouth, with support from the Community Health Council, University of Otago and others, reflecting their commitment to working together to improve the contribution of primary and community care to the wider Southern health system.

The vision for primary and community care is ‘excellent primary and community care that empowers people in our diverse communities to live well, stay well, get well and die well, through integrated ways of working, rapid learning and effective use of technology.’

The strategy has strategic goals to support the vision focusing on empowering consumers, whanau and communities; integrating care across primary, community and secondary care; and a technology enabled health system.

Priority action areas for the delivery of the strategy set out in the action plan were new models of care/workstreams and enabling infrastructures.

Key models of care/workstreams include:

Healthcare home implementation: A patient-centred approach, which aims to combine the traditional core values of general practice with building the capacity and capability of general practice through the development of new roles, skills and ways of working.

Community health hubs implementation: Establishing facilities where secondary outpatient services, advanced primary care services, at least one general practice operate in the healthcare Home model, diagnostic services and other independent and community based healthcare providers work together in an integrated way.

Locality network implementation: Advisory networks made of health professionals and consumers which help to prioritise and plan health services to better align with the needs of local communities.

Home team (rapid response and enablement service): A patient-centred initiative, which aims to help support patients at home via an inter-professional team after leaving the hospital, or a support service at home to avoid to hospital admission. The target group is elderly people.

Consumer-lediIntegrated care: A programme of care to people with long-term conditions using care planning and risk stratification to access more care and provide greater control over managing patient health conditions.

The enabling infrastructures include governance and leadership of the system, health and business intelligence to support planning, funding and delivery; workforce capability and culture; and funding and contracting arrangements to support integrated ways of working.

SDHB, Southern District Health Board.

In 2019, the University of Otago and the Alliance South received funding to evaluate the implementation of the PCCS as a University-Health Sector collaborative project. This article reports on an evaluation that aimed to explore PCCS implementation at the alliance governance level. More specifically, the evaluation aimed to identify facilitators or/and barriers to the successful implementation of the PCCS using a commonly used implementation science theory: the Consolidated Framework for Implementation Research (CFIR)19 20 (see box 3).

Box 3

Consolidated Framework for Implementation Research (CFIR)

The CFIR is a theoretical framework that provides a structure for identifying facilitators and barriers to implementation.19 It offers a comprehensive, standardised list of constructs that allow researchers to identify variables that are most relevant to a particular intervention. The CFIR comprises five domains, namely: intervention characteristics (eight constructs), outer setting (four constructs), inner setting (five constructs), characteristics of the individuals involved (five constructs) and the process of implementation (four constructs). The CFIR has been widely used to inform qualitative process evaluations across a range of complex interventions, including healthcare redesign, in healthcare systems.20

Methods

Design, study setting and sampling

This research used qualitative methods within a pragmatic paradigm, which has a focus on producing actionable knowledge.21 In terms of researcher positionality three members of the research team were academic researchers external to the alliance (GG, TS and CJ), one researcher had previously chaired the alliance ALT (RG). Semistructured interviews were conducted between March 2021 and August 2021 with key informants who were sampled purposively. These constituted the members of the ALT, including a former senior member of the SDHB involved in commissioning the PCCS and the service project leads of the workstreams (see box 3). Interviews were conducted either by video conferencing (Zoom) or face to face. Interviews varied in length from 30 to 45 min. Written informed consent was obtained from all participants.

Data collection

A semistructured interview guide (online supplemental file 1) was used for interviews based on the PCCS goals, a literature review and discussion within the research team. The topic guide covered the main areas of governance team focus (structure, capability and internal relationships), implementation and monitoring of the strategy, perceived impact of the strategy and barriers and facilitators to implementation. Revision and refinement of the topic guide was undertaken as the interviews progressed.

Data analysis

All interviews (undertaken by GG) were audiorecorded and/or videorecorded. Field notes for all interviews were taken, which were expanded by listening to the audio recording and reading automatic transcripts obtained from Zoom.22 23 We used Gale et al’s framework-guided deductive rapid analysis approach,24 which was specifically developed for use with the CFIR.24 25 First, a template summary table was developed in MS Word guided by study research questions and topic guide questions (GG, TS and CJ). This template table was used to summarise individual interview field notes. Next, a matrix in MS Excel by participant type was prepared to chart and consolidate the interview responses in a matrix using the information from the summary table. This process of charting data in a matrix26 was helpful in comparing and contrasting the findings within an individual interview across the different CFIR domains, constructs and subconstructs and between the different participants. This allowed us to categorise the facilitators and barriers of PCCS implementation.

In order to ensure evaluative rigour we used the Consolidated criteria for Reporting Qualitative research27 to structure reporting of the methods and the findings (online supplemental file 2) provides more detail relating to reflexivity, study design and analysis).

Patient and public involvement

No patients or the public were involved in the design of this study.

Results

We interviewed 11 participants. Five were ALT members (chief executive of SDHB, chief executive of Well South PHO, Chief Māori Health Strategy and Improvement Officer, a district Mayor and a community representative). Six were other key informants based on their involvement in the development or implementation of the strategy (former chair of the ALT, former commissioner of the SDHB, three service project leaders of the workstreams—one from the PHO and two from the SDHB—and the executive director for the PCCS).

A number of implementation issues were identified in three CFIR domains (online supplemental file 3): intervention characteristics, inner setting and implementation process (see box 4).

Box 4

Consolidated Framework for Implementation Research (CFIR) domains and constructs

I. Intervention characteristics

Intervention source

Evidence strength and quality

Relative advantage

Adaptability

Trialability

Complexity

Design quality and packaging

Cost

II. Outer setting

Patient needs and resources

Cosmopolitanism

Peer pressure

External policy and incentives

III. Inner setting

Structural characteristics

Networks and communications

Culture

Implementation climate

Readiness for implementation

IV. Characteristics of individuals

Knowledge and beliefs about the intervention

Self-efficacy

Individual stage of change

Individual identification with organisation

Other personal attributes

V. Implementation process

Planning

Engaging

Executing

Reflecting and evaluating

  • Domains and constructs used in this study are in bold.

Implementation characteristics

Complexity

Complexity is defined as the perceived difficulty of implementation, which is reflected by duration, scope, radicalness, disruptiveness and intricacy.19 Participants considered the PCCS a complex intervention and challenging to implement in a short time frame. The sheer magnitude of the action plan required the integration of primary and secondary care. Related to this, some objectives in the strategy were less tangible without explicit activities and milestones.

[The] strategy and action plan is a massive piece of work. …it took a long time to unpack and figure out exactly who was supposed to be responsible for doing different things within the strategy and action plan, because it had such a massive span. It wasn't just one department, it’s like all of primary and community, and then also needed buy-in from secondary to actually make it work. So it was a whole of system approach that required everyone to get on board. (P10)

Another source of complexity was related to the interdependencies of the regional health context. For example, the successful implementation of the Community Health Hub depended on the development of the new Dunedin hospital, a significant project located in the regional metropolitan centre. Therefore, for the success of such projects, there was a need to work out the interdependencies, which appeared to be missing from the strategy implementation.

…if I think about [Community] Health Hubs we've got the things that are going to impact on the success of Health Hubs are outside of the strategy. So things like the development of the new Dunedin hospital, where there is a whole stream of work around, what’s going to exist outside of the new Dunedin hospital, and what’s going to be in an ambulatory care centre and who’s responsible for facilitating that is not clear… (P1)

Inner setting

The ‘inner setting’ is defined as the structural and cultural contexts through which the implementation process occurs.19 Networks and communications, culture, implementation climate and implementation readiness were the constructs identified in this domain.

Networks and communications

The CFIR defines this construct as the nature and quality of social networks, and the nature and quality of formal and informal communications within an organisation.19 Two subthemes were identified in this construct: relationships between individuals, and communication of vision and mission. First, there was a mechanism in place for the ALT members to meet and communicate regularly. Good working relationships at all levels were considered an essential factor in implementing the strategy activities that needed integrated ways of working. However, participants mentioned that relationships between individuals, especially at the higher level, were problematic. This was seen as adversely affecting the quality of communication, hampering open discussion, teamwork, collaboration and feedback. Personalities, personal agenda and power politics, especially at the senior leadership level, were perceived to be conflictual, a barrier to relationship building and contributed to a low trust environment. Participants reflected that conflicts between individuals needed to be managed or addressed early. Otherwise, this leads to an environment of low or no trust and the whole work programme suffers. Second, participants also mentioned that expectations and vision were not clearly communicated to the team members.

…the leadership at the DHB and the leadership at the PHO and individuals didn't necessarily have a good working relationship. That would have an impact on how well the Alliance functions, because I guess it’s hard for people to speak up and have good constructive conversations if key leaders at the table aren't always behaving in that way. (P7)

Culture

The CFIR considers culture as a stable, less tangible and socially constructed idea with the existence of varying definitions. Broadly, it is a given organisation’s norms, values and basic assumptions.19 The governance group (ALT) was expected to facilitate the implementation of the strategy. However, participants highlighted an existing siloed organisational culture between the DHB and the PHO and between primary and secondary care providers characterised by a low level of trust and poor working relationships. The ALT mechanism largely failed to bridge these institutional silos. The silo mentality was not conducive to joint working, collaboration and shared decision making. Related to this, the primary and secondary care sectors had distinct corporate cultures, different scopes of practice and funding models. There was no mechanism in place to facilitate the collaboration between these sectors.

Southern health system is pretty dysfunctional in terms of how the two parts of the health system [the hospital and primary care] work together. There is very little working together between hospital and general practice and community care here. That is been historically the case here, and then at the management level, it became complex and confrontational. You cannot make a change in a health system unless there is trust and confidence between the players. (P6)

Implementation climate

This refers to the collective influence of organisations' policies and practices to promote effective implementation.28 Important issues around the implementation climate that were not appropriately addressed were compatibility and relative priority.

Regarding compatibility, some of the elements of the strategy appeared to conflict with the best interests of some health providers. Participants highlighted that there would be negative financial and workforce implications for general practices if they were integrated into the community health hub model of care, which aimed to move care from hospitals to communities. Therefore, there was a need to identify and manage such conflicts and competing priorities.

There are different drivers for behaviours which are easy to point out, but given that New Zealand is configured in a model of private enterprise delivering primary care and public sector delivering secondary care how do you create solutions and incentives that recognise those very real needs that are different? We have to recognise that the model needs to address those financial incentives and disincentives in a way that makes it to all players. Moving forward in the health system, we have to figure out how we can get processes aligned, recognising there are different competing priorities. (P8)

Regarding the relative priority of this work, the project leaders had to do this strategic work on top of their day job for the organisation they worked for (DHB or PHO). There were no dedicated resources and team to lead individual projects. If something important and urgent came up, this had to be prioritised. The impact of COVID-19 from 2020 also impacted their capacity to make this work a priority. PCSS implementation had to take second place to operational matters.

…when things got really busy and other things came over the top like Covid, so often things would get delayed and the implementation of the strategy was nice to do and went on the back burner. (P1)

Implementation readiness

The CFIR defines implementation readiness as an organisational commitment to its decision to implement an intervention. Two subconstructs were important here: leadership engagement and available resources.19

Leadership

Leadership engagement was about leaders’ commitment, involvement and accountability for the implementation.19 We identified leadership commitment and leadership stability as subcategories of this construct. Participants perceived that a high level of commitment was needed from the leadership of both organisations to resourcing and joint working.

Participants highlighted a need for a shared vision and goal, and for this to happen, the working relationship of key people in both organisations required expert relationship facilitation. Participants also felt that regardless of organisational structure, it was essential to have influential leaders who could work together, identify and devote resources and develop trust and confidence.

You don't need an Alliance to implement the primary community care strategy. You don't need a DHB and a PHO to implement the primary community care strategy. What you need is people who are in positions of influence and leadership to agree, work together, pool their resources, develop trust and confidence to do the work. (P6)

Another issue was with leadership stability. With frequent changes in the key personnel related to strategy implementation, particularly in the DHB, there was a loss of institutional knowledge and momentum of individual workstreams and overall strategy implementation, which was frustrating for staff tasked with delivering the workstreams.

From an implementation perspective, it was incredibly frustrating for people at the grassroots level when there’s such a high number of personnel changes at the top and everyone wanted a different way of doing things. I feel like there was a bit of a stumbling block to the implementation. (P10)

Resources

Poor resourcing was a major barrier, as many workstreams were not adequately resourced. Projects with dedicated resources and a change team were successful in meeting their objectives. For example, the healthcare home was relatively successful with dedicated resources, leadership and project management of the PHO.29 In contrast, the community health hub and locality networks had no such resource in place and competing priorities, which was reflected in slow progress. Participants were critical of assumptions that care could be shifted from the secondary to the primary sector without extra funding and human resources.

…it [the Strategy] generated a lot of expectations and a lot of work, which were not adequately resourced. What we ended up with was a whole lot of things on the action plan, which didn't have a lot of resources allocated to them. And it was to be done on top of your day job type thing. And it became very difficult. (P2)

… if you stopped doing something in a hospital space, how does the primary care workforce just do it magically for no extra cost? So in other places in the country where things that are moving out of the hospital and to the community, there’s a funding stream for that. But trying to get those funding streams in place in Southern has been so difficult…(P6)

The main reason participants offered for poor resourcing was that the ALT team members representing the SDHB and the PHO could not agree on the distribution and assignment of available resources.

While the ALT was supposed to facilitate SDHB and PHO support for the PCCS implementation, they had no authority and no mandate for decision-making and budget/resource allocation so they only could advise instead of directing.

So the Alliance didn't have any authority. They had all authority over being able to advise [as opposed to direct] the PHO and the DHB. So they don't have any formal mandate and they don't have any delegation [delegated authority]. There’s no budget, but they do advise us on things in relation to primary community strategy but it wasn't decision-making as such. So it was almost like an extra layer. (P1)

The perceived funding crisis in the health system, with large SDHB budget deficits and ‘no spare money,’ also played a role, as did a perceived tension over which organisation should hold budgetary authority.

Implementation process

The implementation process includes four interrelated activities essential for successful implementation: planning, engaging, executing and reflecting and evaluating.19 During the strategy development, implementation was planned to occur in stages mainly due to the financial cost to the DHB and PHO and associated workforce implications. In hindsight, it was evident that there was not enough preparation to implement the strategy. Participants identified a number of issues across all four processes, which are discussed below.

Planning

Participants mentioned that the strategy and its action plan included the list of activities (what components) but lacked details on how to implement the strategy. There was a lack of clarity around roles and responsibilities, mandate and scope in some of the essential components of the strategy (community health hub and locality networks). There was a strategy and action plan, but a detailed implementation plan which provided enough direction for the execution of the strategy implementation was lacking. A need for a single project management approach with a shared vision and goal was also highlighted.

…there was no roadmap for implementation. It was only a very high-level set of things in the action plan. (P4)

It’s [implementation of the strategy] not just like turning on a tap and suddenly everything is in place. We ended up with a strategy, a framework that was to be progressively implemented by Southern DHB and WellSouth. It was a strategy that was always going to be implemented in stages and progressively partly because of financial cost to both the DHB and the WellSouth, but more importantly, because of the change in workforce practice that was a consequence of the strategy. (P5)

Engagement

Engaging with implementation leaders, stakeholders and community people, including vulnerable groups is vital for the successful adoption of the change process and to ensure the needs and interests of various groups are addressed.19 The PCCS required engagement with stakeholders while developing the strategy and its implementation. As mentioned earlier, the strategy implementation was complex requiring engagement with multiple stakeholders to secure buy-in and ensure successful implementation.

Mixed feelings were expressed about engagement while developing the strategy. Different stakeholders such as GPs, Māori (iwi groups), the Clinical Council and community groups across Otago and Southland were consulted. However, a few participants highlighted that a broader range of community groups could have been consulted.

Participants also perceived that the strategy was not developed by following a bottom-up process engaging with staff from SDHB and the PHO. It was felt that the process could have been better with more robust engagement with staff and leaders of both organisations.

… the primary community care strategy was written with the help of consultants, and so there was an element of it not being really well built from the ground up. (P6)

Executing

This component relates to carrying out or accomplishing the implementation according to plan. Most of the participants agreed that the strategy’s implementation had not been realised as per the initial vision of the strategy. Overall, they perceived that except for a few projects, the execution component could have benefited from greater planning. As mentioned above, the success of the execution was hampered by the lack of resources, clarity in scope and structure of the projects and a detailed implementation plan. Furthermore, the projects’ interdependencies were not fully understood at the outset, so the implementation of headline activities (models of care) remained an isolated effort. As a result, there were variations in the progress of elements of the action plan. The healthcare home was seen as having been successfully implemented. In contrast, with no resources and no detailed plan, community health Hub and locality networks implementation progressed slowly.

For the different components, it [execution] is different. You know the Health Care Homes have gone really, really well. Community Health Hub has really struggled. The locality networks need some more energy. (P8)

It’s an ongoing theme in health and in the public sector, which is we've got fantastic ideas and we're good at analysing the problem and we're good at knowing what we need to do, but we're not so good at the implementation piece. (P7)

Reflection and feedback about the progress and quality of implementation is an important way to promote shared learning and improvements along the way. There was good reporting to the Alliance team regarding updates about some workstreams (primary maternity, healthcare home, home team, etc). However, participants mentioned that their reports to the broader programme group appeared to be more of a compliance activity rather than an improvement. Participants suggested a need for an overall quality improvement approach in place for the whole strategy.

There was a lack of an overall quality improvement approach or method. How about using a continuous quality improvement approach across the Alliance structure, down into the individual projects? (P4)

Participants felt that the feedback mechanisms could have been more robust, and it was difficult for them to contribute to decision making at the governance level.

Some of the instruction coming from ALT was a little bit hard to understand. The SIC [Service Improvement Committee] group did feel a little bit unsure about what expectations were or what feedback was required and didn't necessarily feel that they were being heard. (P2)

Participants highlighted that there was a lack of a clear structure and performance measurement framework to measure and track the progress for many strategy objectives.

Discussion

Summary of findings

Our findings provide insights into the experience of an Alliance and senior health professionals of the overall implementation of the PCCS, which aimed to promote integration between primary and secondary care. We found that the large number of strategy action plans and interdependencies of activities made implementation of the strategy complex. Communication and relationships between individuals and organisations were identified as an important factor for joint and integrated working, but needed a more favourable environment than that of the pre-existing organisational silos. Essential elements of a positive implementation climate were not adequately addressed to better align the interests of health providers, and there were multiple competing priorities for the project leaders. A perceived low level of commitment from the leadership of both organisations to joint working and resourcing indicated poor organisational readiness. The combination of no detailed implementation plan with a single project management approach, ambitious targets, the lack of a clear performance measurement framework, and an inadequate feedback mechanism demonstrated several gaps in the implementation process and resulted in poor execution of the strategy.

Strength and limitations

The use of the CFIR strengthened the evaluation by offering comprehensiveness, capturing the dynamics of the implementation process’ complexity, systematising the analysis process and helping to tell the story by organising and producing rapid actionable evaluation findings.20 30 31 The collaborative nature of the research was also helpful in facilitating the sharing of findings and feedback to the Southern health system. We further used a rapid analysis approach23–25 32 that was both efficient in terms of researcher resources25 and allowed us to provide prompt feedback to the local health system in line with our aim to provide actionable knowledge. A limitation of using such a rapid approach was that we did not use detailed transcription and line-by-line open coding process, which might risk missing nuances of data. A further potential limitation of our approach is that we did not formally feedback our findings to each participant (respondent validation). Nonetheless, we shared our preliminary findings in periodic colloquiums targeted to participants from the Southern health system and the feedback we received was that they ‘made sense’ to participants.

While the use of the CFIR helped us to illuminate contextual factors and understand the interplay between the context and implementation process of the complex strategy intervention, it would be helpful to identify how different constructs identified by the CFIR interacted to produce certain outcomes. A further issue is that ‘complexity’ as used by the CFIR is defined through an implementation science framework as a complex (complicated) intervention; this is different from the way ‘complexity’ is framed in health systems research that uses complexity theory.33 In this regard, the way CFIR structures its implementation process constructs (from planning to evaluating) is also linear, though CFIR’s developers acknowledge the process may follow a non-linear course.19 Although beyond the scope of our study, one possible future approach could be to combine the CFIR with a realist informed evaluation.34 This would enable CFIR identified constructs to be used to identify mechanisms for how different contextual conditions generate outcomes.35 It is interesting to note that our analysis does suggest underlying mechanisms (eg, lack of trust, poor leadership) that impact negatively on achieving the desired outcome of PCCS implementation.

Comparison with existing literature

We used the CFIR to understand the implementation of the PCCS in the NZ southern health system context. Our study adds to a small body of research using implementation science theory (CFIR) to study health system transformation initiatives.20 36–38 Using the CFIR, the study was able to unpack the black box of complex relationships between the intervention, its context and the implementation process.

Our study also adds to the body of NZ health services research addressing what conditions underpin successful implementation of integrated care. Previous studies into alliancing aimed at integrating primary and secondary care have also identified the issues reported here as being key to implementation success.8 10 13 38 39 The Canterbury initiative of delivering integrated health and social care using alliancing highlighted a number of key enablers: the development of a clear and shared strategic vision, continuity of senior leadership, staff engagement, a continuous quality improvement approach and development of new ways of contracting for health services.13 39 Another NZ evaluation of pilot initiatives using alliancing to promote primary and secondary care integration reported overly ambitious plans, competing priorities, inadequate attention to organisational culture and lack of timely funding support as barriers to implementation.38 Similarly, an evaluation of the system level measures framework found that the strength of formal and informal organisational relationships at the local level were critical conditions for implementation success.10

The findings are also consistent with the international literature on alliancing.14 16 Experience of US Accountable Care Organisations emphasised the importance of having realistic expectations, finding ways to develop trust, managing conflict and making a collective decision, and focusing on leadership.16 Our findings echo previous studies in the UK about similar health system transformation initiatives, most notably Vanguard programmes.14 15 40–42 Vanguards were local pilot sites established across England to develop and deliver National Health Service (NHS) initiated new care models to coordinate care across primary care, community services and hospitals.14 15 These studies highlighted the need for realistic expectations, for local capacity and capability building, the importance of developing relationships, strong leadership, robust data and analytics and managing time constraints.14 15 40–42

Implications for health policy and practice

A number of important lessons and potential solutions useful for future health policy and practice that have emerged from this implementation evaluation are summarised in table 1.

Table 1

Key lessons learnt and potential solutions

One key finding of this study is the importance of leadership—for successful local health system change a stable and committed senior leadership team working through high trust relationships and open communication across all partner organisations is needed. Previous health system transformation initiatives in NZ and internationally also offer valuable lessons relevant to integrated care through alliancing.13 It is noteworthy that while the alliancing approach in the NZ health sector was initiated more than a decade ago9 13 the lessons from the Canterbury initiative have not been fully realised across NZ. One reason for this is likely to be a failure to contextualise the alliancing approach for different local health systems.37 Another is the possible ongoing influence of managerialism as a guiding principle for organisation. This influence may be to the detriment of leadership development, especially clinical leadership, which has been shown to be beneficial for healthcare organisations and developing new models of care.43 44

The NZ government’s health sector reforms, enacted in mid-2022, focus on addressing the problems the health system is facing, including service fragmentation.7 The reforms emphasise the need for better integration between primary and secondary care.7 The reforms create a single NHS through a new structure called Health New Zealand and have a greater emphasis on working with local communities through geographical locality networks with aims of integrating primary and secondary care.7 45–47 Alliance South had been in abeyance since the announcement of the new health system reforms in April 202146 and, at the time of writing, the details of the structure of governance and service delivery at the local level were in development. Regardless of the new configuration of the NZ health system, we consider that the lessons learnt from this implementation evaluation will be instrumental for planning and implementing future initiatives at the local level.

Conclusion

Using the CFIR, this study identified factors for the successful implementation of the PCSS using an alliancing approach in Southern NZ. During the evaluation period, wide-ranging health sector reforms in NZ were announced. Those leading the reforms should consider the key lessons from this study—in particular the importance of a stable and committed senior leadership team working through high trust relationships and open communication across all partner organisations—to strengthen integrated primary BEand community care delivery, which are core reform goals.

Data availability statement

No data are available. Full deidentified interview transcripts will not be shared. Informed consent, in line with the approving ethics committee, only allows for the use of deidentified extracts within research reporting and writing to maintain participants’ privacy based in a defined regional area and population, thus making their identification with full transcripts more likely.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by University of Otago Human Ethics Committee. Reference number: D21/044. Participants gave informed consent to participate in the study before taking part.

Acknowledgments

The authors would like to thank the participants who gave their time for interviews. We are particularly grateful to the Southern Primary and Community Care Strategy Evaluation Steering Group (Chair, Dr Carol Atmore) for their support. We also thank Lisa Gestro, Executive Director Strategy, Primary and Community, Southern DHB for her input during proposal development.

References

Supplementary materials

Footnotes

  • Twitter @StokesTim63

  • Contributors TS, CJ and GG conceived and designed the study. GG conducted interviews and led the data analysis with input from TS, CJ and RG. GG, TS and CJ drafted the manuscript. All authors read, provided critical reviews and approved the final manuscript. TS is the guarantor.

  • Funding This study was funded by Health Research South (https://www.otago.ac.nz/oms/research/hrs/) (2018 Health Service Delivery Research Grant; 'Evaluating the Implementation of the Southern Primary Care and Community Care Strategy' award/grant number: N/A).

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.