Health Care Home implementation in Otago and Southland: a qualitative evaluation

Keywords: CFIR, evaluation, health care, Health Care Home (HCH), implementation or implementation science, model of care, primary care, qualitative research.

Health systems internationally^1,2 have developed new models of primary care to address a range of health-care challenges, including increasing patient complexity, increasing demand, workforce shortages and financial constraints.³ The Health Care Home (HCH) is one such model. It was first developed in the United States (US) in the 2000s as the patient-centred medical home (PCMH).^4,5 The PCMH is a patient-centred approach, which aims to combine the traditional core values of primary care – comprehensive, coordinated, integrated, quality care that is accessible and based on an ongoing relationship between patients and their primary health-care practitioners⁶ – with building the capacity and capability of primary care through the development of new roles, skills, and ways of working. The US PCMH has been shown to improve outcomes in terms of quality, cost and utilisation.⁷ Canada implemented elements of the PCMH in its early 2000s primary care reforms,⁸ developing the Patient’s Medical Home (PMH) model, which also has an explicit focus on community adapativeness and social accountability.⁹ Variations of the PCMH have been implemented in Australia^10,11 and the UK.¹²

In New Zealand (NZ), the HCH (Table 1)¹³ was developed by Pinnacle Midlands Health Network (a Primary Health Organisation (PHO)) in 2011.^3,14,15 There has subsequently been roll out of the HCH model across NZ¹⁴ and, to date, over 200 general practices across NZ have or are implementing the HCH.¹⁶ Uptake of the HCH has been voluntary, with general practices being supported since 2016 by the Health Care Home Collaborative (a membership network of PHOs and District Health Boards (DHBs)).³ Despite this uptake, there has been limited evaluation of the NZ HCH both in terms of its effectiveness¹⁴ in improving health outcomes and whether the model can be successfully replicated in different health-care settings to those in which it was initially developed.^14,17

Table 1.  The HCH model of care.¹³

In NZ’s Southern Health District (Otago and Southland), the Southern DHB and the WellSouth Primary Health Network (PHO) are working together to implement the Southern Primary and Community Care Strategy (PCCS), which seeks to ensure all ‘live well, stay well, get well’.^18,19 The PCCS aims to develop a sustainable Southern health system for the future. In 2019, Southern DHB and WellSouth PHO partnered with the University of Otago’s Centre for Health Systems and Technology (CHeST) to set up an academic–service partnership to evaluate the implementation of PCCS and to build evaluation capacity within the Southern Health System.

One PCCS key initiative is the HCH. HCH implementation started in the Southern district in 2018, and by July 2019, it had been rolled out in 15 practices (first wave). Implementation support to practices was provided by the PHO HCH implementation team for 3 years and included regular visits to participating practices. Our evaluation aim was to explore the facilitators and barriers to HCH implementation in the Southern district using a commonly used implementation science theory: the Consolidated Framework for Implementation Research (CFIR)^20–22 (Box 1).

Semi-structured interviews were conducted between July 2020 and March 2021 with key informants from first wave HCH practices in the Southern district (Otago and Southland). To maximise variation, practices were sampled purposively to include a wide range of populations and profiles. The key informants comprised general practitioners (GPs), nurses and administrative staff. Interviews were conducted either by video conferencing (Zoom) or face-to-face. Interviews varied in length from 30 min to 1 h 45 min.

The interviews used a semi-structured topic guide (Supplementary material File 1) based on a literature review and discussion within the research team. The topic guide covered the implementation of key HCH components, including barriers and facilitators to implementation. The interview guide was used flexibly to allow participants to construct their accounts in their own terms and was revised and refined as the interview progressed.

All interviews (undertaken by GG) were audio and video recorded. Field notes for all interviews were taken, which was expanded and completed by reading automatic transcripts generated by Zoom and listening to the audio recording.^23,24 A rapid thematic analysis using a framework method^25,26 was conducted by GG and TS. This was informed by the CFIR.^20,21 First, a summary template was developed in Microsoft Word (Microsoft Corporation) from the recorded responses guided by study research questions and topic guide questions.²⁷ Through an iterative process, this template was used to code the rest of the field notes.^24,27,28 Then, the CFIR domains and constructs were applied to the responses to categorise implementation issues. Next, a data display matrix in an Microsoft Excel spreadsheet was prepared to chart and summarise the interview responses in a matrix using the completed summary templates.²⁵

In addition, key HCH practice documentation (reports and implementation plans) were reviewed to supplement the interview data. The data were analysed using a qualitative content analysis approach.²⁹ The coding framework (summary template and data display matrix) used in interview transcripts was applied to the content of the document.

Ethics approval was obtained from the University of Otago’s Human Ethics Committee (D19/295).

We interviewed 15 participants (13 individual and two group interviews) from seven HCH practices (two GPs, six nurses, and seven administrative staff). Of the 15 participants, 13 were female and two were male. Follow-up interviews were conducted with six participants (one individual and two group interviews) from two practices (practice C and F) after 7 months to capture changes over time. The practices varied by geographic location, population characteristics and practice size³⁰ (Table 2).

Table 2.  Participating southern region HCH practices.

Barriers and facilitators to HCH implementation were identified across three CFIR domains (Box 1). Illustrative participant quotes are presented.

This domain emphasises the importance of the need to adapt interventions to enhance their fit with the context. The CFIR conceptualises an intervention has core components (the essential elements of the intervention) and adaptable periphery (adaptable elements, structure and systems related to the intervention and the organisation into which it is being implemented).

Although the intervention – the HCH model – had well-designed core components, participants viewed the need to consider the adaptable periphery while implementing core elements such as GP triage and patient portals. Participants observed that modification (adaptation) was needed for when and how to use GP triage according to practice needs and contexts (eg practice size, enrolled population, staff). Practices with a reasonable clinical capacity to manage urgent demand did not need to implement GP triage daily. It was found that these issues were addressed as the implementation progressed. Similarly, the existing funding model for general practice was described as a barrier to fully realise the benefits of the HCH, indicating that the current fee-for-service model was not suitable for clinicians to see patients who need longer appointments. The HCH transformation calls for changes in routine operations of practices. Hence, overall emphasis should be given to changes in structure and practice processes (adaptable structure and systems) to fit the HCH elements into practices.

PHO is very keen for us to do all day GP triage, but because we have quite good [appointment] availability here, we use it only when we’re fully booked. And that has worked really well. (P13, Practice Manager)

Some practices had already introduced a number of changes similar to the HCH concept before formally adopting the HCH model. Such prior implementation experience was a positive implementation factor.

The team’s experience of the patient portal leading to improved efficiency, decreased workload and greater patient satisfaction means they are already seeing the benefits of an HCH approach. (Document review)

Furthermore, evidence from published literature and anecdotal stories regarding the HCH implementation in other regions of NZ was also considered a positive factor for adopting the HCH model in their practices.

The ‘inner setting’ is defined as the structural and cultural contexts through which the implementation process occurs. Implementation climate and readiness for implementation were the major constructs identified in this domain and are reported here.

Implementation climate refers to the collective influence of organisations’ policies and practices to promote effective intervention implementation. We identified the key CFIR sub-constructs as tension for change and compatibility.

A perceived need for change in practices was a positive factor in the HCH implementation. Some practices were preparing for the change process to manage acute demand, and they found the introduction of the HCH tools such as GP triage very timely to address this issue.

We were trying to work through some processes and put in place some systems to help manage that volume [number of patients presenting to practice to be seen on any given day]. And the Health Care Home actually came along at a perfect time… (P1, Nurse Practitioner)

Participants mentioned that the HCH model did not work well if multiple owners shared the business and held differing views. Another compatibility issue was possible tension with commission-based GP remuneration where practices pay GPs commission on the income they generate. Because HCH tries to get people to not be seen in person if they do not have to, there is the chance that GPs may consider they may lose income if they see fewer people.

You can’t do Health Care Home as a practice if you’re three different owners sharing a premise. And you cannot do this when you have GPs working for you on a fee for service basis. It’s not going to work. (P11, Practice Manager)

Readiness for implementation was discussed in terms of the CFIR sub-constructs of leadership, available resources and access to knowledge and information.

Participants noted that strong leadership in practices with organisational commitment was critical in driving change. Here, the leadership team encouraged and engaged with team members to adapt to and accommodate change and to try new things. Furthermore, it was important for the leadership team to find time to meet, work together, and communicate decisions with the rest of the practice staff. An important dimension of organisational commitment is managerial patience, which was essential for implementing changes. The HCH change process needs initial investment costs and time, although it pays off in the long run. For example, when new tools such as GP triage and portal were introduced, these initiatives increased workload on staff, particularly nurses and doctors. Hence, it is necessary for organisational commitment from the leadership until the change process starts to have an effect.

So making the time for everyone to come together and implement this model of care for us is sometimes a real barrier. It’s important to have a GP, a nurse, a practice manager, and admin manager all working on it. However, sometimes that doesn’t always work and it falls back on one person to lead it. (P15, Practice Manager)

Participants highlighted that funding support from the PHO HCH implementation team was very helpful, which allowed practices some funded time to understand the HCH process and benefits well.

The Health Care Home funding has allowed us to allocate time to stop being busy to look at other ways [of doing things] … (P1, Nurse Practitioner, Follow-up interview)

Participants also mentioned that practices that managed their available (internal) resources in terms of funds and skills enabled them to implement and manage changes quickly; however, some practices, particularly smaller practices, found it challenging to manage resources (eg human resources, time and structural capabilities).

Participants highlighted that adequate orientation and training related to HCH tools and change management and timely practical support for implementing different HCH tools were necessary.

So we’re all kind of doing the same work which could have been done by WellSouth which would have then been disseminated to potentially every practice in the South Island. (P3, Practice Manager)

Related to this, participants described confusion and resistance from staff, particularly GPs and nurses, during the initial stage of implementation in almost all practices. Patients had little awareness of the changes made to the practice’s services as a result of adopting the HCH model. Access to knowledge and information to both patients and staff around changes made in the system was therefore felt to be essential.

… patients didn’t really understand what was going on. And I don’t think most of them even realise now what Health Care Home really is. (P10, Practice Nurse)

CFIR identifies the process by which the implementation is accomplished as requiring four activities: planning; engaging; executing; and reflecting and evaluating.

All practices had a detailed implementation plan with clearly developed targets and activities, which meant clarity of purpose on what they were going to do and how they were going to achieve it. However, participants viewed that too many changes in the first year were overwhelming to staff and practices. Hence, doing things in stages was thought to be important. Furthermore, practices need a long-term plan beyond the initial 3 years to sustain the changes.

Participants mentioned that adopting the HCH model required a large number of changes that would take time to realise their benefits. It required a shift from reactive (GP-based) to proactive care (team-based). It was also reported that implementing practical components such as GP triage and the patient portal was straightforward, but the most challenging thing identified was change management: these transformation efforts needed an effective change management process.

…things like GP triage, patient portal are easy, tangible things that you can do and invest, but the hardest nut to crack is change management… (P1, Nurse Practitioner, Follow-up interview)

As the HCH model means a transformation to a new way of service delivery, participants believed that whole-of-practice engagement was crucial. This was, however, felt to be missing in some practices. Engaging team members right from the beginning and educating the entire practice team about the HCH model was important in getting buy-in from the team. Working collaboratively and justifying to the team members why certain changes were important were crucial to getting support from the practice team.

There’s a lot of changes happened in the first year. It was quite overwhelming for the team. We realised that we need to slow down and involve people right from the start. (P13, Practice Manager)

Participants considered there was a need for change in the whole system by shifting more resources and care of delivery to primary care to fully realise the HCH vision, and that this required engagement with secondary care (Southern DHB).

A further aspect of engaging identified by participants was the use of external change agents – the PHO’s HCH implementation team. Participants highlighted the need for ongoing practical support from this team to address ongoing issues in the HCH practices.

The Health Care Home implementation or continuation team has to still exist to provide practical support to practices to say, well, this practice had this problem, this is how they solved it… (P2, GP, Follow-up interview)

HCH components that appear to have received more implementation focus by the practices were GP triage, daily huddles, patient portal and LEAN principles. As noted above, it was not realistically achievable to implement all the HCH components in the first 12 months. Coronavirus disease 2019 (COVID-19) generally slowed HCH implementation as the staff and resources were diverted to manage it. Nonetheless, COVID-19 prompted moving forward with other parts of the HCH model, including telephone consults, huddles, and electronic prescriptions.

The programme had a clear matrix of performance indicators and a monthly report from practices that allowed assessment of each practice’s HCH progress. Furthermore, the PHO HCH team provided feedback to practices during practice visits. Practices also used a self-assessment maturity matrix to track their progress. Overall, the PHO HCH team emphasised a continuous quality improvement approach to improve the implementation.

Secondly, participants emphasised the importance of sharing knowledge with other practices to learn from each other and avoid reinventing the wheel. Some participants mentioned that they met monthly to share ideas and achievements.

…with the Health Care Home, it’s really good that we meet once a month and practices share information with each other. Everybody’s different, but you learn so much more from each other. (P11, Practice Manager)

Finally, participants were pleased with the PHO HCH team’s approach to working with practices, which was seen as being collaborative (joint learning and journey together) rather than a compliance activity.

Total collaboration. That has been awesome. It’s probably the first time we’ve done anything that it hasn’t been pushed on you. We’ve kind of done the journey, learning… (P13, Practice Manager)

This study explored the implementation of the HCH model in first-wave HCH practices in Otago and Southland using implementation science theory (CFIR). Implementation facilitators and barriers across three CFIR domains – intervention characteristics, inner setting and implementation process – were identified. The HCH model had well-designed core components, but more attention is needed to the adaptable periphery to tailor the programme elements for the local context. Implementation climate and readiness for implementation were the constructs identified in the inner setting. A successful implementation climate requires a strong need for the change (tension for change) and compatibility. Strong leadership and availability of resources, among others, were important implementation readiness factors. A staged implementation, change management plan, whole-of-practice engagement, and importance of sharing knowledge and information among practices were the most important facilitators of a successful implementation process.

The study has a number of strengths. First, interview data were collected from a diverse range of participants, and was complemented with document review, which helped triangulate the findings. Second, we have identified no other NZ study using an explicit implementation science framework (CFIR) to understand HCH implementation. The CFIR helped to systematise analysis and organisation of findings, and capture the complexity of the implementation process. Third, this was a university–health sector collaborative research with one of the investigators (SB) holding an active leadership role in supporting practices to roll out the HCH initiatives. The collaborative nature of the research was also helpful to facilitate the sharing of findings and feedback to the Southern health system. A further strength is the use of a rapid analysis approach^24,28,31 to provide prompt feedback to the local health system.

A key limitation was that the recruitment of participants was challenging, as we could not include two practices that we originally planned. COVID-19 also affected the recruitment process for a short period, as the practices were occupied with pandemic work. We did not use detailed transcription and a line-by-line open-coding process, which might risk missing nuances of data. However, we did use automated transcription generated by Zoom and listened to audio/video recordings multiple times to complete field notes and help categorise participant responses across themes.

In terms of comparing our findings against the international HCH implementation literature (US³² and Australia^10,11), one important caveat is that the health system context in which the HCH is implemented varies widely and may limit transferability of findings to the NZ context.³ For example, PCMH development in the US occurred within a health system heavily focused on specialism and hospital-based care,³³ and a key barrier in the US PCMH literature, lack of an integrated electronic health record,^10,32 is not present in NZ. Nonetheless, the international literature also finds that engaged leadership is a facilitator.^10,11,32 Similarly, in line with our findings, difficulties with change management, funding models, insufficient internal resources (a particular problem for small practices) and staff engagement are barriers.^10,11,32

There is a limited NZ literature on HCH implementation, consisting of grey literature, and none has been published in peer-reviewed journals.^34–37 This literature, though it does not use the CFIR, also highlights organisational readiness for change as an important facilitator, and resistance to change (in the initial implementation phase) as an important barrier.³⁴ One barrier found in the NZ literature, not identified in this study, was high staff turnover.³⁴

The HCH model of care is currently being rolled out to further general practices in the Southern district. Feedback of these context-specific implementation findings would be of benefit to these new HCH adopters, and have been fed back to the PHO HCH implementation team. Of particular note is that practice size matters; smaller practices will likely have a lack of internal resources, necessitating greater external (PHO) implementation support. Our key findings are likely transferable to other NZ health regions. Our identification of factors facilitating and impeding HCH implementation by using CFIR domains should be helpful for planners and managers to better identify the implementation gaps.

This evaluation identified a number of implementation facilitators and barriers across three CFIR domains: intervention (HCH model) characteristics, inner setting and implementation process. Existing and future HCH practices need to consider these factors when implementing changes. It is imperative to tailor the HCH model to local needs and individual general practices for successful implementation.

Full de-identified interview transcripts will not be shared. Informed consent, in line with the approving ethics committee, only allows for the use of de-identified extracts within research reporting and writing, in order to maintain the privacy of participants who are based in a defined regional area and population, thus making their identification with full transcripts more likely.

This study was funded by Health Research South (https://www.otago.ac.nz/oms/research/hrs/) (2018 Health Service Delivery Grant; ‘Evaluating the Implementation of the Southern Primary Care and Community Care Strategy’).

Tim Stokes is an Editor of the Journal of Primary Health Care, but was blinded from the peer review process for this paper. Stuart Barson is a contractor to WellSouth PHO and Marc Haughey is an employee of WellSouth PHO.

Supplementary material is available online.