A rapid evaluation of staff and patient experiences of care delivery as part of a vertically integrated model of care Open Access

NHS policy in England over the past 10 years has repeatedly stressed an objective of more integrated patient care across primary and secondary health care and social care settings (NHS England, 2014, 2016, 2019, 2022). The policy focus on care integration led to the development of recommendations in the 2014 NHS “Five Year Forward View” (NHS England, 2014). In England, the NHS Five Year Forward View described, among other options for stronger integration between primary and secondary care, what it termed Primary and Acute Care Systems (PACS), which would combine general practice and hospital services as a type of accountable care organisation (Maniatopoulos et al., 2020). A report from The King’s Fund recommended that acute hospitals should play a greater role in primary care provision whereby acute hospitals work more closely with primary care providers developing integrated service models that span organisational boundaries (Naylor and Charles, 2018).

Since 2015 a number of local initiatives, rather than centrally mandated policies, have led to NHS trusts running GP practices. This is a form of vertical integration: where one organisation provides care at different stages along the patient pathway (Rumbold and Shaw, 2010). Vertical integration may include differing levels of integration across a health care system to provide coordinated, patient-centred services. The focus of the work reported here is integration across primary (general medical practitioner (GP)) and secondary (acute hospital) care. Vertical integration can make decision making, monitoring and information sharing more efficient. In health care, this might translate as better mutual understanding between GPs and hospital specialists; better communication and flows of patient information between them and reduction of risks through better demand management (Conrad and Dowling, 1990).

An important rationale for vertical integration in the NHS has been to enable general practices to remain open. Other reasons for vertical integration in the NHS include: expectation of efficiency improvements, to address pertinent public health concerns and to improve co-ordination across acute hospital and general practice services (Sidhu et al., 2022). Key to achieving vertical integration is better clinical integration (coordination of treatment services for a patient) and functional integration (strengthening support functions such as financial management, human resources and strategic planning). Our statistical analysis reported elsewhere (Sidhu et al., 2023), as well as that of Yu and colleagues (2020), finds that vertical integration between providers of secondary care (including acute hospitals) and local GP practices leads to modest but statistically significant reductions in the rate of A&E attendances, unplanned hospital admissions and readmissions (Sidhu et al., 2023).

Similar integration of some general practice and community services with hospital care has also happened, for example, in Spain and the USA (Comendeiro-Maaløe et al., 2019; Schwartz et al., 2018). The rationales driving the implementation of vertical integration appear to vary country to country but are not dissimilar to those observed in the UK and included: the expectation of efficiency improvements, integrating health care to address pertinent public health concerns and improving co-ordination across acute and general practice services. Concerns have been raised about applying learning from international vertically integrated models of care to the UK For example, the NHS Confederation stated that widespread roll-out of vertical integration in UK could lead to local commissioners having less control over which services are delivered locally to patients and could lead to some providers of care being squeezed out (NHS Confederation, 2012).

A review by Baxter et al. (2018) suggests that the implementation of integrated care models in general – that is not limited to vertical integration – may lead to improved patient satisfaction and perceived quality of care. Patients appreciate more seamless access to services, while staff report that integrated systems can enhance communication and collaboration across disciplines. Baxter and colleagues find that while organisational changes aimed at integrating care services in the UK show potential benefits in terms of patient satisfaction, the success of such initiatives largely depends on effective collaboration and communication across primary and secondary care settings.

Despite previous research on vertical models of integration in the UK and internationally there is little evidence on staff and patient experiences in the specific context of vertical integration. We address the research questions:

1. What impact is vertical integration having on the patient journey with regard to access to and overall experience of care?

2. How do models of vertical integration support patient transitions from primary care to acute care?

We undertook a qualitative, cross-comparative case study, mixed-methods evaluation (Sidhu et al., 2023). We engaged with theoretical frameworks and concepts to guide our interpretation of data. First, we used the SELFIE (Sustainable integrated care modeLs for multimorbidity: delivery, FInancing and performancE) framework (Leijten et al., 2018) to guide the development of our research questions and inform our interview questions. The SELFIE framework structures concepts and features of how best to design and deliver integrated care for patients living with multiple long-term conditions. It uses a holistic, person-centred approach incorporating broader socio-economic, behavioural and community-level factors. Other frameworks, such as the Chronic Care Model (Wagner et al., 1996), we find less well-suited for our study than the person-centred considerations emphasised in SELFIE. Having completed data collection with stakeholders, we inductively identified themes and saw the relevance of concepts from the sociology of work (Turnbull et al., 2019; Corbin and Strauss, 1985) and from theories of relational coordination (Gittell, 2006).

We draw on qualitative data from: (1) focus groups and semi-structured, one-to-one interviews with clinical and non-clinical staff from acute hospitals and GP practices; and (2) one-to-one interviews capturing the views of patients (and their carers) from integrated GP practices.

We undertook comparative case studies of three vertical integration sites across England (April to May 2022). The research team identified NHS Trusts in England as of 31 March 2021 where there was vertical integration between an acute hospital and GP practices. We took a purposive approach to identify a sample of case study sites with variation in: (1) geographical location and population served; (2) legal and governance working frameworks and (3) time elapsed since vertical integration was introduced. Table 1 summarises the case study sites.

Focus groups and one-to-one interviews were conducted with a purposive sample of staff. Interviewees included chief executives and directors (clinical and non-clinical) and other trust managerial staff (related to integration and strategy, delivery of health care services, and financial and governance-related management). We also interviewed GPs and other primary care staff who have implemented the vertical integration model in each area.

For interviews with patients we sought participants living with multiple long-term conditions. They tend to have frequent contact with general practice services and hence are particularly exposed to the consequences of any changes in care provision. Members of the study team (IL, MS and GM) spoke with leads at Primary Care Networks (PCNs) across each site to engage with members from Patient Participation Groups, using the National Institute for Health and Clinical Excellence (NICE) definition of multi-morbidity as an inclusion/exclusion criterion (NICE, 2016).

The salient characteristics of staff and patients interviewed across our three case study sites is provided in Table 2.

Members of the study team met with members of a Patient and Public Involvement (PPI) group to discuss the study; the questions that focus groups and interviews with staff and patients might explore and methods of recruitment to ensure inclusivity (90-min online meeting in May 2022). Participant-facing documents were reviewed by PPI group members and their feedback was incorporated. A further online meeting with PPI group members was held during data analysis and write-up to share learning and cross-check authors’ interpretations (90-min in November 2022).

Interview and focus group participants were approached by local clinical leads/managers of vertical integration models from each site. If agreeable, they were then contacted by a researcher who sent them an information sheet and consent form, which participants were asked to return ahead of their interview/focus groups. Focus groups were completed in person while interviews were conducted over Microsoft Teams, Zoom or telephone. Focus groups lasted 45–90 min. Interviews lasted 20–90 min. Focus groups and interviews were audio recorded (subject to consent), transcribed verbatim by a professional transcription service, anonymised and stored in compliance with the General Data Protection Regulation (GDPR) 2018 and Data Protection Act 2018.

Data collection and analysis were carried out in parallel and facilitated through a sub-team of researchers (MS, IL, GM and FW) holding bi-monthly, one-hour telephone meetings for the duration of data collection, analysis and write-up (June to December 2022). Three analysis workshops were held by the team from September to November 2022 to discuss data in the context of findings from all case study sites. Our work was informed by the Gale et al. (2013) framework for the analysis of qualitative data in multi-disciplinary health research. NVivo 12 was used to aid inductive thematic analysis. All transcripts (N = 19) were independently thematically coded (three transcripts were coded in duplicate to increase rigour in our coding) to validate interpretations from early online workshops. The coding work continued until researchers (MS, GM and FW) were unable to develop any further emerging themes, that is saturation was reached (Ritchie et al., 2013). (A detailed description of the full coding structure is presented in Appendix 3 of Sidhu et al., 2023).

We then undertook a second cycle of analysis that was conceptually informed by the notion of the sociology of “work”, as described in Corbin and Strauss’s seminal paper on three lines of work (illness work, everyday life work and biographical work) when managing chronic illness (Corbin and Strauss, 1985; Turnbull et al., 2019). We used the theory of relational coordination to interpret how staff across case study sites described their process of developing their respective vertical integration model (Gittell, 2006). The theory of relational coordination is based on three dimensions of work relationships: (1) shared goals; (2) shared knowledge or understanding and (3) mutual respect and four dimensions of communication (timely, frequent, accurate and problem-solving communication).

In total, the study team approached 26 members of staff to complete an interview across all three sites. Twenty-two agreed to take part (Greenvale = 9; Townshire = 6; Urbanville = 7). Nine participants took part in three focus groups, and 13 completed one-to-one interviews. The study team approached 24 patients to complete an interview across all three sites. Fourteen agreed to take part (Greenvale = 2, Townshire = 4; Urbanville = 8). Despite the moderate response, authors generated a varied sample regarding high and low users of primary and secondary care, gender, ethnicity, age and range of long-term conditions.

We identified drivers at the meso- and macro-organisational levels that influenced the development of vertical integration, which in turn determined patient pathways and experiences at a micro level. The fundamental rationale for vertical integration in Greenvale and Townshire was to sustain primary care local to where patients live. It is likely that without support from the local acute trust many of the practices which integrated would have closed. The principal rationale for vertical integration in Urbanville was for the acute trust to work more closely with primary care to improve local population health outcomes. Senior managers and executives from acute trusts, from all three case study sites felt vertical integration can improve the sustainability of general practice and improve services local to where patients live.

In Townshire, staff from the acute trust reported facing challenges from their local commissioner, previously the Clinical Commissioning Group (CCG) and now the Integrated Care System (ICS) and Integrated Care Board (ICB). These were apprehensive about vertical integration supporting general practice in their area, a feeling that staff understood did not extend to other large-scale providers of primary care in the area:

Senior primary care clinicians from Greenvale remained unsure what impact the local ICS and ICB would have upon their vertical integration model. Senior staff in Urbanville thought the ICS was unwilling to engage and understand how the acute trust was attempting to break boundaries between primary and secondary care to provide more co-ordinated care:

Acute hospital and primary care staff from all three case study sites recognised that vertical integration can establish a model where the acute trust acts as an anchor organisation, which can then work at scale with primary care networks across general practice. Yet, there was an organisational dichotomy between the hierarchical governance structure introduced by acute trusts to establish vertically integrated models, compared to the flatter structures found between general practices integrating horizontally within primary care networks (PCNs).

This dichotomy led to tensions within PCNs that included both vertically integrated and not vertically integrated practices concerning how care should be delivered in the local health economy. In Townshire, where the acute trust is working with both vertically integrated and not vertically integrated practices as part of the same PCN, an managerial member of staff from primary care cited difficulties in taking opportunities for care-ordination:

In Urbanville, by contrast, staff at the hospital trust felt the PCN was an effective meso-level organisational model to establish relationships with general practices and understand the challenges that primary care clinicians face. A PCN in Urbanville that consisted solely of vertically integrated practices was used as a test bed where service innovations could be piloted and, if successful, rolled out to neighbouring PCNs.

Given that national policymakers and ICBs are moving towards a more network-based governance model for general practices, the degree of success of vertically integrated models in improving coordination of care and hence patient experience and outcomes will depend on how acute trusts deal with tensions around funding and service delivery.

To achieve service improvements for patients, acute trust and GP leadership across the sites described the value of establishing and developing working relationships between the acute trust and the general practices in the local area. Interviewees described two priorities: (1) establishing trust between senior leaders in general practices and the acute trust; and (2) senior acute trust managers and clinicians understanding the complexities of delivering primary care, its funding mechanisms and how best to re-organise its service provision.

A major challenge during early stages of vertical integration was to address perceptions that the acute trust was an “avaricious takeover beast” (A2, Secondary Care, Clinical and Managerial) and show GPs that it had the interests of primary care in mind:

One senior clinician from Townshire acute trust illustrated how shared understanding of each other’s operational intricacies meant more realistic expectations of the other and more considerate working relationships:

However, a few clinicians felt that the cultural and organisational gap between primary and secondary remains too large to bridge. Hence, beyond administrative efficiencies there were only limited service provision improvements that could be realised by vertical integration between primary and acute secondary care.

All three case study sites displayed elements of innovation affecting patient care: new/improved service delivery, reimagined ways of working and new technologies. Both staff and patients we interviewed felt innovation was better facilitated in a vertically integrated model. The most significant component that was attributable to vertically integrated working was acute trusts providing pump-priming funding for “innovation projects” that would benefit all general practices locally. GPs in Urbanville described changes to health care delivery as “innovation projects”, which were conceptually categorised in two groups: process and structure. Process innovation projects implement changes to internal working practices to better support the co-ordination of healthcare delivery. Structure innovation projects change internal or external infrastructure and thereby enable new methods of working.

One example of a process change was stratifying the patient population and conducting analyses to assess patient needs across the health economy. Structural innovations included the recruitment of GPs who worked across both vertically integrated and not vertically integrated practices in the area, a pharmacy hub that managed repeat prescriptions to alleviate GP burden and an integrated centralised telephone system to manage appointments:

In Townshire, staff described the introduction of structural innovations, including a community midwifery team located in the same building with GPs. However, Townshire encountered several challenges to introducing innovations, notably from the local commissioners. One senior manager in primary care reported that an initiative to introduce a renal clinic with consultant input failed to materialise due to concerns from the ICB about equity of access for patients:

But overall, it appears that structural and process innovation projects are effective at the local level through collaborative working between primary and community care staff, with oversight from clinicians in secondary care.

Across all three case study sites we found issues with patients’ access to primary care services, but these were not exclusive to vertically integrated practices, they were common across primary care in general. Patients were offered various modes of access to primary care including, but not limited to, in-person, by telephone, and digital via apps and online. This led to significant navigation work being undertaken by patients to access primary care services. Many patients resorted to in-person techniques to obtain urgent appointments while using digital media for routine care. Overall, patients felt access to general practice was challenging:

Across the three case study sites, patient satisfaction with access to secondary care was mixed. At Townshire and Greenvale, patients perceived improvements to the transition of their care between primary cand secondary care following the introduction of vertical integration. Alongside the introduction of specialist clinics outside hospital – e.g. for physiotherapy – treatment was good once patients were seen in secondary care:

Yet there were other accounts of patients experiencing long waits to be seen and to have a procedure despite vertical integration:

These findings are within the context that the landscape of navigation and illness work shifted considerably for many patients during the COVID pandemic, regardless of the presence or absence of vertical integration. Many patients switched to accessing care via remote methods: communicating with clinicians in primary and secondary care via telephone, online platforms and email.

Vertical integration is a model of integrated care that can help general practices remain open. It also provides opportunities to introduce novel ways of delivering clinically integrated care across primary and secondary care settings aimed at improving patient access to and experience of health care. However, although particular service innovations may be tried in vertically integrated practices first, they are not then limited to those general practices but rather are rolled out more widely. Findings from our study show that vertical integration supports embedding clinicians from secondary and community care into primary care settings through multi-disciplinary working. In addition, the roll-out of “innovation projects” across our case study sites demonstrates the level of flexibility and autonomy that vertically integrated trusts and practices have to deliver new services on their own initiative. We note that following the Covid-19 pandemic and the spread of new ways to access health services, patients and especially those with multiple long-term conditions, encounter significant “navigation work” choosing and accessing health provision. But this is a general change, common to both vertically integrated and not vertically integrated primary care practices.

Our findings extend the literature on understanding the drivers for vertical integration and its impact on the patient journey and experience across primary and secondary care. Given there are no national policy guidelines on the implementation and monitoring of vertical integration in England to compare against, our manuscript provides a novel case study-based depiction demonstrating bottom-up vertical integration initiatives in response to addressing local health system challenges as opposed to a top-down mandated policy requirement.

Shortell et al. (2000) identified several challenges that need to be addressed to achieve better integration of care. Our findings are consistent with Shortell and colleagues’ assessment. At the case study sites we found that the adoption of vertical integration was in part a response to financial realities threatening the closure of some general practices. Misunderstandings by some primary care staff of the objectives of acute trusts had to be overcome, and the historical organisational separation in the NHS between primary and secondary care meant that mutual trust and understanding had to be built so that there was collective buy-in to integration. We found considerable engagement between GPs and secondary care clinicians and managers at the case study sites in attempting to understand how vertical integration could support and alleviate pressures faced in primary care.

Robinson and Casalino (1996) highlighted that vertical integration focuses on integration between acute care and primary care entities, due to expected advantages that include improved transitions across care settings, as well as facilitating co-operation where financial incentives across settings are not aligned. We found a similar focus of vertical integration at our case study sites.

We build on the notion, described by Turnbull et al. (2019), of navigation work encountered by patients and their sense-making strategies and help-seeking behaviours. We have applied this notion when interpreting patient journeys as part of vertically integrated models of care. We found that patients continued to face challenges – continued to undertake navigation work – when accessing care despite vertical integration.

The study team completed a rapid, qualitative, comparative evaluation, following established methodology while iteratively engaging with published literature. We spoke to many staff and patients (13 interviews and 3 focus groups with 22 staff; 14 patient interviews) and are confident we achieved data saturation.

However, at two of the three case study sites fewer than intended staff and patient interviews were completed. This was due to workforce challenges in general practice and moderate interest from the patient participation groups we contacted. We recognise that our sample of patients interviewed cannot be taken as representative of the population of England more widely. Our sample was more informed than patients generally would be about the model of vertical integration introduced in their area. Patient interviews were constrained in what they could reveal about the exact contribution of vertical integration to patients’ experiences. Finally, as part of this study we were unable to compare patients’ accounts between those registered with vertically integrated practices and those with non-integrated general practices in the same local area as vertically integrated practices.

Implementation of vertical integration and subsequent innovations in patient care occurs in stages, which may happen in parallel or sequentially: (1) contractual and legal changes for acute trusts to take responsibility for general practices’ contracts with the NHS, either directly or via a subsidiary company; (2) organisational and functional integration; (3) cultural understanding- and trust-building amongst colleagues from secondary and primary care; and (4) developing and introducing innovative practices to improve patient care and pathways.

Following a sustained period of embedding that includes addressing system-level changes to improve organisational and functional integration, vertically integrated models have shown success in establishing collaborative working across health sectors locally to bring about service improvements. Staff from acute hospitals were more likely to achieve changes to clinically integrated patient pathways with a bottom-up, facilitative approach built on trust with colleagues from primary care.

Further research is particularly needed on patient experiences of navigating and receiving care as part of vertically integrated models, including patients with multiple long-term conditions, but also those living with complex conditions that require continued clinical monitoring and patients managing long-term mental health conditions. Such research should include the views of carers supporting patients.

At the three case study sites, we find that vertical integration supports the sustainability of general practices while fostering clinically integrated working between primary and secondary care to better meet local patient needs. Successful implementation requires an initial phase of trust- and relationship-building between clinicians and staff across both sectors. However, even with vertical integration, patients continue to face significant navigation work in choosing and accessing healthcare services.

Funding sources: This is independent research funded by the National Institute for Health and Care Research, Health and Social Care Delivery Research programme (BRACE Project no. 16/138/31). The views expressed in this publication are those of the authors and not necessarily those of the National Institute for Health and Care Research or the Department of Health and Social Care.