Key points

Question

What factors enable or hinder the successful implementation of Hospital at Home (HaH) services in England, particularly in response to NHS England’s 2021 policy promoting ‘virtual wards’?

Findings

Five key themes influenced HaH implementation across eight services and three Integrated Care Systems:

  1. Adaptation of clinical practice: Success required clinicians to modify care processes, electronic records use, and interdisciplinary roles for home-based settings.

  2. Change agents and leadership: Effective leadership and prior experience in innovation were critical for overcoming implementation challenges.

  3. Access to resources and support: Funding, technology, staffing, and senior managerial support enabled service growth, though inconsistent access created barriers.

  4. Governance structures: Supportive, flexible governance enabled innovation, while rigid or unclear policies led to service tensions and delays.

  5. Shared learning and stakeholder engagement: Collaboration within MDTs and ongoing dialogue with hospital-based clinicians helped integrate and expand HaH pathways.

Meaning

To scale HaH services effectively, health systems must invest in adaptive clinical practices, distributed leadership, flexible governance, and structured learning environments. Policymakers should recognize HaH as a complex innovation that requires time, resources, and context-sensitive approaches to embed and sustain. Early funding helped, but long-term scale-up needs systemic support beyond initial implementation.

Introduction

Hospital at Home delivers acute care in the patient’s usual place of residence. There is growing evidence of its efficacy, primarily focused on clinical and cost effectiveness.1 In 2022 the world Hospital at Home community released research priorities for HaH, including the identification of enablers and barriers for implementation and scaling of services.2

In 2021 NHS England (NHSE) implemented a policy encouraging all Integrated Care Systems (ICSs) to establish ‘virtual ward’ services,3 however little is known about the factors at service level and the wider context that influences implementation. Virtual Wards were initially conceived as a technology-supported remote monitoring system,4 rather than delivering acute multidisciplinary care at home. However, the NHS England 2024 operational framework has renamed these services ‘Hospital at Home’ and changed the specification so that the care model is closer to the internationally recognised HaH definition.5 This change of emphasis, provides a unique opportunity to study the implementation of services at a various of stages of maturity (see Figure 1).

Figure 1
Figure 1.Representation of Policy Implementation Over Time

A 2024 Cochrane review found 52 qualitative studies in the literature on barriers and facilitators of HaH services, identifying four main themes6: development of stakeholder relationships; processes, resources and skills required for implementation; acceptability and caregiver impacts; and sustainability of services. It was noted that all studies had methodological concerns, and that 16 were subacute or rehabilitation-based care which would not meet the world Hospital at Home definition.7 Furthermore, the review noted a lack of evidence at a healthcare system level (i.e. at an organisational level above that of direct care providers), mostly being individual services reviews, and none were carried out in response to the reform of a national healthcare management body – the main mechanism through which large scale future HaH programmes are likely to be delivered.

In order to fill the gap outline by the Cochrane review we set out to study system level (through the ICBs) and provider level (through individual sites) of HaH. This study seeks to understand the service and system-level factors influencing the implementation of HaH services in response to NHS England’s virtual ward policy

Methods

We used a comparative qualitative case study design, which is a methodology for studying health policy implementation and change practices within real world contexts.

Our case studies comprised eight HaH services (described as services in this paper) across three integrated care systems (described as systems in this paper) which reflected inner city and more rural geographies (see supplemental information for details). An ICS is a local health partnership in England consisting of health and social care providers and commissioners responsible for NHSE policy implement. Sites were selected to provide diversity and complexity of context comparison8 and to understand how different system and service contexts and organising processes influence implementing ‘virtual ward’ and HaH services. We used multiple approaches to data collection, including observations of everyday activities, interviews and documentary sources, in accordance with established qualitative research techniques.9

Semi structured interviews10 were undertaken with 81 participants including clinicians (medics, nurses and allied health professionals), policy makers and managers (see Table 1) between 2023-2024. Participants were initially purposively sampled then ‘snowball’ sampling was used to identify additional participants. Participants provided written consent to take part in this study. A topic guide was developed and subsequently refined to sensitise it to arising themes. Interviews were audio recorded and transcribed. We undertook observation for all services, including site visits, referral processes, multidisciplinary team (MDT) meetings, documentation and clinical care processes. Observations were recorded through detailed notes that were written up, in full, within 24 hours.11 Documents collected included national and local policies, service standard operating procedures (SOPs) and patient feedback.

Table 1.Professional Details of Participants
Participant profession Number of interviews
Policy maker 13
Manager 15
Physicians 19
Nurse or Nursing associate 19
Allied Health Professional 15

A thematic analysis12 of our data progressed through three partially overlapping stages, involving: i) data coding; ii) theme identification and; iii) theme refinement and definition. Coding was performed independently by 2 of the researchers to identify initial themes, then discussed and refined by the wider research team. A rigorous approach was employed through continuous triangulation across our multiple data sources (i.e. interviews, documents and observations) and searching for repeated occurrences and/or absences of events and categories, across places, times, people and situations.9,13 Participant feedback on themes was sought through a workshop at the end of the data collection.

Ethics approval was granted through the University Warwick (ethics committee approval reference number: HSSREC 61/22-23)

Findings

The services included in our study covered a range of clinical specialties including frailty, respiratory, paediatrics, palliative care and general surgery. Sites provided three main models of care: face to face HaH; hybrid face to face combined with remote monitoring; and remote monitoring only with reliance on other teams or hospital admission for face to face review. All sites were consistent with the NHSE 2021 operational policy guidance,4 although services represented a variety of service configurations and levels of care which is consistent with other studies14

We found five main themes that impacted implementation (see Table 2): clinical practice adaptions; change agents and leadership; access to resource and support; governance; shared learning and dialogue with clinical stakeholders.

Table 2.Summary of Themes with Short Description
Themes Brief summary of enablers Brief summary of barriers
Clinical Practice adaption Clinical practice within the HaH setting is complex and requires clinicians to undertake a series of practice adaptions in order to deliver clinical care, while maintaining professional standards. Clinicians able to take this into account found implementation easier. Services that were unable to support their clinicians to adapt, or clinicians that struggled to adapt found implementation more challenging
Change agents and leadership Leaders with prior experience of change management or practice adaptions found it easier to overcome implementation challenges Leaders with limited change management experience or motivation for the HaH model progressed service development more slowly
Access to resource and support Services were able to implement more rapidly where there were readily available clinical resources (including staff and technologies) and support from senior management Services struggled to implement or scale services where resources (including staff and technologies) were scarce or not fit for purpose
Governance Governance structures at a system level that supported service flexibility or innovation were found to be enabling Where governance structures focussed on performance or financial measures, clinician engagement and service progression was more difficult
Shared learning and dialogue with clinical stakeholders On-going dialogue between HaH and hospital based clinicians encouraged referrals and built confidence in the new model. Where shared learning opportunities were scarce, services found implementation to be slower. Problematic or absence of dialogue delayed service implementation

Clinical practice adaptions

Implementation of HaH services were enabled where MDTs were open to adapting clinical practices to support acute care delivery in home settings. This involved innovating new care models through a complex range of change management activities that simultaneously supported delivering acute care in a geographically dispersed way and accommodating multiprofessional practice standards. We have grouped these into three domains which relate to the professional standards common to healthcare professionals: clinical care process; electronic patient records (EPR) and documentation; and interdisciplinarity.

Clinical care processes – successful adaptation

Clinical care processes, defined as the way of delivering care, were often different or outside the traditional scope of the professional. Clinicians able to improvise and adapt their clinical practice to the home setting, found it easier to deliver care e.g. by modifying processes to undertake clinical examinations or observations, or using remote monitoring or video calls. Clinicians described adapting to the challenges of delivering care, where the equipment may not be set up for clinical assessment. In order to undertake an ultrasound one clinician needed to remove the arm of the chair the patient was seated in to have sufficient access, another clinician described using a head torch to increase the light for placing venous access devices (see Table 3) and others utilised household items such as picture hooks or a lamp to replace a drip stand.

Clinical care processes – challenges to adaptation

Where services were unable to adapt their clinical practice, either due to perceived risk or objection to the clinical model, they relied on a traditional hospital based approach to deliver care. This included bringing patients into the hospital for physical assessments or to receive medications deemed too risky to be administered at home (see Table 3).

Electronic Patient Records (EPR) and Documentation – successful adaptation

There are no current standards for EPR within HaH in the NHS and in particular there are no recommendations to avoid duplication of recording. New technologies or ways of documenting care were embraced to ensure services maintained their governance standards. Successful services worked to overcome challenges with existing EPR systems ensuring suitability for use by community and hospital based clinicians. One service had worked closely with their IT department to produce a version of their EPR that could be used on a tablet or smartphone so that clinicians in the home could input directly into it, and hospital based clinicians could view their documentation remotely.

Electronic Patient Records (EPR) and Documentation – challenges to adaptation

Even with some modifications to their patient records systems, duplication was observed across multiple systems. One service was required to document on a community and hospital based EPR system, and on their remote monitoring platform. Information could not easily be extracted for reporting requirements which resulted in further duplication of recording to support mandated reporting by NHS England. Where these challenges were not successfully addressed, one service had to stop treating patients until a resolution could be found (see Table 3).

Interdisciplinarity – successful adaptation

MDTs typically included a range of physcians, nurses and allied health professionals. We found that the roles of clinicians within the MDT were modified so services could work efficiently. Clinicians described taking on roles that would not traditionally be in the scope of their profession. For example a consultant medic described assisting a nursing colleague with a patient’s personal hygiene during a visit (see Table 3). Shared clinical leadership and responsibility for patients across the full MDT was commonly seen. Trust was placed in the visiting clinicians by senior decision makers. These multidisciplinary relationships supported shared learning and collective innovation within services.

Interdisciplinarity – challenges to adaptation

Implementation was impeded when MDT members were averse to adapting practices to support interdisciplinarity. For example, where services did not have the provision for medical review at home we observed sustained disagreements in MDTs around the clinical responsibilities for patient care. In these settings, clinicians repeatedly sited lack of confidence in a model where a consultant was not physically seeing the patient but was still held legally responsible for care delivery.

Change agents and clinical leadership

Leaders of hospital at home services came from a variety of clinical backgrounds. Where implementation was more established or progressing more rapidly, leaders and/or change agents had prior experience of developing new services or adapting their professional practices to new clinical practice environments . This was not always within acute care or HaH services. These leaders were able to draw on previous experiences to navigate challenges in the implementation processes. They also described a clear vision of the service they were trying to achieve, and were willing to be flexible in their approach.

Governance

Governance structures varied between systems and we found examples of multi-provider collaboratives, system wide steering groups and ad hoc support. These arrangements tended to focus on operational or financial performance of services at a system level. Governance systems struggled to understand the variation in models and often focused on encouraging standardisation of practice across geographical areas. This led to tensions between systems and services around assessments of operational performance (see Table 3). Where system support was more ad hoc, services felt they had more freedom to provide services in a model suited to their local needs and resources. Mostly system level implementation support and governance were subsumed into existing managerial roles.

Although NHSE provided policy guidance to support governance structures, the policy documents were changed during the period of this study. While policy makers and most system managers stated they found the guidance to be clear, clinicians found multiple ways to interpret the guidance to ensure they met the required standards, as evidenced by the variety of service models encountered within this study.

Access to resource and support

Access to resources and support were essential for services to progress their programme. We have divided this into three domains: centralised funding; technology; and staffing and senior support.

Centralised funding

The NHS England two year funding programme was widely viewed as facilitating service implementation. Centralised funding supported service providers to recruit staff, expand teams and introduce technological and clinical developments, allowing services to grow and establish in a short timeframe. However in some situations the flow of money from systems to services was complex (see Table 3). Services experienced challenges with the temporary nature of some funding leading to problems with staff recruitment or training to the appropriate level. Despite funding being available, some services reported difficulty in recruiting suitably qualified staff due to the number of services being set up simultaneously. Difficulties also occurred within systems containing multiple hospital trusts and one service had to stop a specific pathway due to funding disagreements between acute and community providers. When the two-year funding programme ended some services reduced bed capacity within their HaH programmes to meet new budgetary requirements.

Technology

Some services were readily able to access remote patient monitoring technologies through system wide schemes, thereby allowing more rapid implementation into their HaH programme. In one system a region-wide remote patient monitoring solution had been procured for all services. Support was also provided with the technology implementation and refining it to suit the local need. Conversely services within this region reported challenges in accessing other technologies, such as point of care testing, as the system had no further budget to support these. Other systems found the region wide procurement opportunities difficult and were unable to agree with other service providers on the specifications required or struggled to find equipment that met their needs (see Table 3).

Access to staffing and senior support

The multidisciplinary team was central to the delivery of all models of HaH care. Access to appropriately trained staff was crucial for all services to increase their capacity. In one system, to overcome recruitment challenges, nursing had been outsourced to a third party provider. This arrangement was an enabler by facilitating nursing staff to deliver care to patients who lived outside this service’s usual geographical service boundaries.

Leaders described support and encouragement from senior management as being important for service development. Services with senior support found it easier to progress innovation and adaption of clinical policy and practice through complex hospital-based committees and governance processes. Where senior managers supported HaH care, this increased access to funding and technologies described above (see Table 3).

Shared learning and dialogue with clinical stakeholders

We found that implementation was supported where services and systems could support HaH MDTs learning to develop new care pathways and service models together. Shared learning systems were also required to help professional service users (e.g. other clinical colleagues working in acute care and community specialties) understand how to make use of these new services for their patient populations. We have organised this section into two domains: shared learning arrangements for the MDT: and clinical engagement and dialogue with clinical stakeholders and service users

Shared learning for the MDT

At service level, individuals created opportunities to learn from each other, e.g. accessing expertise from more established sites through site visits, and established professional interest groups sharing learning together. Clinicians described supporting learning within the MDT as a key role in HaH. One clinician described supporting a range of MDT members to learn to use point-of-care ultrasound, a task previously limited to medical staff.

Systems were keen to support shared learning and we found a variety of activities including events and meetings to share learning internally. One system was supporting provider development sessions, where services could feed back on new pathways being developed and have an opportunity to share learning, although clinical attendance at these was variable. For wider collaboration, systems relied heavily on the NHS Futures platform which had a range of webinars during the early stages of policy implementation and a platform for sharing documents. Sharing of knowledge through the implementation stage was not usually clinically led and was mostly ad hoc.

Dialogue with clinical stakeholders

Clinicians described needing to maintain an ongoing dialogue with their hospital-based colleagues about their service and newly developed pathways of care. This increased confidence in referrals to the HaH pathways and helped to bring patients into the service. Furthermore this dialogue reduced isolated of clinicians from their hospital based colleagues while adapting clinical practices for home based care.

Leaders developed relationships with key stakeholders and were persuasive when they met resistance to their implementation goals. A service leader described meeting regularly with consultants and specialist nurses from a specific specialty to allay fears that the HaH service was taking over their roles. They described changing their conversational style depending on who they were talking to and ultimately bringing them into the service where they now work together to support patients jointly.

Service implementation was impeded when service leaders had difficulties engaging clinical specialties and building relationships to overcome colleague concerns. Several services described ambitions to develop new care pathways being prevented by concerns from clinical colleagues about increasing workload and perceptions of increased risks (see Table 3).

Table 3.Description of Themes with Representative Enabling and Disabling Quotes.
Theme description - enabling Representative quote Theme description – disabling Representative quote
Clinical practice adaptions
Clinical practice within the HaH setting is complex and requires clinicians to undertake a series of practice adaptions in order to deliver clinical care, while maintaining professional standards
Clinical care processes
Services need to support clinicians to adapt their clinical processes to deliver care in the home		 “I think sometimes we have to do things slightly different in the community and… You know, like, hanging a drip from a, a coat hanger on a curtain rail cos you don't have a drip stand” – service clinician Clinical care processes
Services that have not adapted clinical care processes find it more difficult to deliver in the home		 “So at the moment if they go out and do a visit, and they’re worried, the only way we can do a clinical assessment is bringing them back to our day case unit for clinical assessment. So we get observations, we can phone the patient if we need to, but if they need a physical examination, or something, then they have to come up to the day case ward for that.” – service clinician
Documentation
Embracing new ways of documentation helped services maintain their governance standards		 “probably the single biggest issue that we've resolved is our documentation. Patient documentation we are recording on an Excel spreadsheet that isn't properly backed up. That doesn't meet information governance standards, and I'm not comfortable running a service with that standard of documentation that to me falls below the line.” – service clinician Documentation
Services that were unable to overcome documentation challenges		 So, we didn’t have any additional staff, and we were using the community team to do home visits, and they use a different clinical system, we use [EPR 1], and they use [EPR2], and the two systems didn’t talk to each other, and so in the end it just didn’t work very well. So we couldn’t see how the patients were doing if we needed a clinical update, because our [EPR1] system was out of date…So, we got a few patients out, and then it wasn’t really up scalable, or, doable on a long term basis, so we stopped it”-Service clinician
Interdisciplinarity
Supporting the roles of the clinical team to change and adapt to deliver efficient care		 “We, we don’t really have that in the hospital. So I think they work in silos, doctors and nurses. But, here they're working together….I think it needs a lot more team working and it's not necessarily that the traditional doctor role wouldn’t work… sometimes the doctor needs to put in the cannula and take the blood as well or Clinical Practitioners also taking on roles that traditionally would be doctor roles” service clinician Interdisciplinarity
Sustained disagreements within the multidisciplinary team effects implementation progress		 “The biggest issue that I think people have, has been the blame, and it's always, it always, always comes down to if that patient dies in a virtual ward, and I haven't seen them, I'll get hauled before the coroner.” – service clinician
Change agents and leaders with prior experience of change management were more able to navigate implementation challenges. “ever since I started there, my role had kind of been to implement a virtual ward. But the virtual ward that we had there 10 years ago was much more about chronic disease management… So, I think I just, you know, since I've come to [Hospital A], I've been really excited about how enthusiastic everybody is to make this work and I'm determined that we will make it work”- service clinician Change agents and leaders with less experience or motivation for the hospital at home model found implementation to be slower “The problem they’ve got at the moment is you’ve got people like me who’ve been doing my job for ten years, who don’t necessarily want to suddenly start doing something completely different, cos you get to a point where, there’s only so many different things you can do.” -Service clinician
System Governance
Governance structures that allowed clinicians freedom to develop services suited to local need gained confidence more quickly “And that's partly why we decided really the acutes [trusts] need to own it and grow it because only they will know. And this is probably about clinicians giving their patients discharge, if you're discharging to yourself or your own respiratory team or frailty team and it's one of your colleagues running that, I'll use the term literally "step down virtual ward" to get them out the actual hospital bed, you're going to have a lot more confidence” – system manager Governance structures that primarily focused on operational metrics found clinician engagement difficult “NHSE targets of 80% [capacity] and 40 to 50 beds [per 100,000 patient population] does not drive quality anywhere in that at all. It drives number crunching and it drives moving patients around and pushing them out. I struggle to see the quality agenda in it and I think that's where our clinicians have struggled as well.” – System manager
Access to resource and support
Access to staffing and support
Services were able to implement more rapidly where they had readily available clinical resources (including staff and technologies) and support from senior management		 “I've obviously, been pushing it forward and I think there's buy in from everybody, really, that it's the right thing to do. So, yeah, I think it's the managers here are on board.”- service clinician Access to staffing and support
Services struggled where access to resources were challenging		 “not everywhere has been able to recruit all of these roles, because although the roles are being created, there aren't the people around to fill them. And I know there's been a little bit of worry about virtual wards taking people from places that are already sparsely populated with staff. So, you're not really getting more people trained out. So, are you moving people from one thing to a virtual ward, which leaves another area short?” – Service manager
Technologies
Access to technologies that supported the clinical care model supported implementation		 “the introduction of technology along the way that’s supported that, so, erm, point of care blood tests, so you can have an immediate result in the patient’s home, rather than having to drive the bloods back to the hospital, send them to the lab, wait for the bloods to come, and then if you need to do something you’ve then got to drive back to the patient, so that adds in quite an inbuilt delay, whereas if you can have the blood test in the patient’s house, you can…, act on the result, and, and be able to do something immediately” – service clinician Technologies
In ability to access suitable technology caused delays in the implementation processes.		 “Because actually getting really good remote technology has been really, really difficult. So, we trialled something and basically it wasn't what it said on the tin…I trialled the thing as well as other ones, so I know exactly what it was. And it was monitoring my temperature, at times, 4° out” – service clinician
Centralised funding supported services to invest in staff and technologies to support implementation “the way that the NHSE and the money has worked is we thought we had a two year, fully funded period piece of work to buy us the time to do this, to fail fast, to try new things, to scale up, gather the evidence.” – system manager Centralised funding
At times the flow of funding to services was complex and caused challenges with service development		 “Because you can say, well, if you expand, if you double your virtual ward how many more staff will you need? Well, I've no idea, but I can guess. So, we put together some numbers, submitted them and then it was, well, you've asked for too much, you need to cut the numbers down. So, we kind of say, well, how much money is there and then we'll work around the money? Oh, no, we can't tell you that, you need to tell us what you need. So, frustrating, there's a lot with finance about what we can and can't afford.” – service clinician
Theme description - enabling Representative quote Theme description – disabling Representative quote
Shared Learning and dialogue with clinical stakeholders
Learning and an ongoing clinical dialogue supports clinical pathway and service development.
Shared learning
Services that shared learning across the MDT found services developed more rapidly		 “ I think, er, my role is, as much about supporting [and training] the team as it is about caring for the patients, if you see what I mean? I think because of having consultant support, the team has developed more”– service clinician Shared learning
Services where the MDT worked separately or did not share learning found services developed more slowly		 “you’d never have an inpatient ward with just being seen once a day by a… nurse and nobody else, that, that is not a ward by any stretch of the imagination. So we have completely hamstrung ourself locally by offering such little MDT input into these patients”- service clinician
Shared learning
Where systems shared learning, implementation progressed more quickly		 “I think sharing and learning the PDSA part of it has been absolutely key to our implementation as well. And we do a lot of that as soon as Virtual Wards go live as part of the provider development session, they do-do the PDSA and review and refine. A lot of shared learning, I think a lot of learning from other areas as well.” – System manager Shared learning
An absence of shared learning or communication around implementation at system level challenges lead to tensions		 So I think, in some places there was a bit of tension… So maybe we could have workshopped it better or, we didn’t workshop it at all, we could have done that, taken the opportunity to engage with people. – system manager
Dialogue with clinical stakeholders
On-going dialogue between hospital at home and hospital based clinicians encouraged referrals and built confidence in the new model.		 “it was really difficult through that first year, 18 months. And there was almost a riot in the grand round about we're not taking on any more work and things. And that's why our frailty consultants have been pretty pivotal in being like the beacons of positivity for it and showing how it can be done.” service clinician Dialogue with clinical stakeholders
Problematic or absence of dialogue at a service level delayed service implementation		 And I suppose that was taken out of my hands by, “You deliver it by Monday.” And I was, like, “That’s not happening.” “Okay, you deliver it by the [date], that’s the last possible day, just do it.” Er, so, so I just did it, and then, and then there’s the challenge of, okay, so I can't get… I can't do this in a perfect way, I just have to try to pull people together” – service clinician
Dialogue with clinical stakeholders
Systems that facilitated dialogue between services and systems found implementation easier		 “So we tried a, we tried a number of different ways, erm, so we brought all of the providers together, and actually, we gave them, they’d done some work looking at, at discharge previously, as a group, and their discharge pathways… So we started to flip the story, we tried to, between us, find who we knew might be more positive that others, brought in, erm, some very positive voices from primary care, some more of the supportive secondary care” – system manager Dialogue with clinical stakeholders
Problematic or absence of dialogue at a system level delayed service implementation		 “I think the, the more, more certain leaders, erm, signed up to the idea that it wasn’t safe, because of the risk, then the more the rest of consultants would, kind of, follow that message. So the messaging from quite high up was that it wasn’t going to be safe.”- system manager

Discussion

We believe that this is the first study in England to look at the barriers and facilitators to implementation of HaH services and adds to a relatively small evidence base globally. Although there have been some studies looking at early support discharge programmes,6 this study is unique and extends the extant evidence base in a number of ways. Firstly, our study considers implementation at both a service and system level. Secondly, our research design included input from a wide range of stakeholders including clinicians, managers and policy makers with an extensive range of experience working across the HaH field of practice. Finally, although our study identified some implementation challenges, we found services implemented new HaH models rapidly and in accordance with the NHS England policies that were in operation at the time.

There have been similar studies examining implementation, barriers and facilitators for HaH globally. Studies from the US, Australia and Singapore have examined single services15,16; focused on patient perceptions17 or stakeholders18; or specialised in residential care19 or rehabilitation.20 However, the context for implementation in England was unique, given the large scale, top-down, directive for services to be set up and the entirely publicly funded health system. Despite these differences in health care systems, similar themes such as leadership and senior support, financial sustainability and stakeholder engagement can be seen throughout the global literature.

Based upon the findings discussed, we have categorised the themes into those that predominantly have their effect at a service or system level. We provide recommendations to support implementation and future service scaling, summarised in Table 4, based upon the key themes identified.

Table 4.Summary of Recommendations to Support Implementation
System level recommendations Service level recommendations
Where possible expand the remit of existing services to deliver hospital at home Set realistic implementation targets incorporating the increased length of time necessary to adapt hospital practice to the home
Create dedicated hospital at home leadership roles rather than adding this to the portfolio of existing leaders Appoint service leaders with a track record of change agency i.e prior experience of innovation or practice adaption
Create long term financial resourcing models for recruitment, training and scaling of services. Ensure change agents have visible support of senior management within their organisation .
Formalise clinically led learning by experience of setting up, developing, and delivering hospital at home Change agents must have access to:
  • peer professional mentoring (e.g. from more experienced implementers);
  • interdisciplinary clinical professional learning collaboratives (e.g. via professional societies or system-wide arrangements);
  • professional leadership development;
  • clinical skills development across full MDT.
Systems must tolerate initial variation amongst providers due to the heterogenous case and starting point of the innovation journey

System-Level Recommendations

At the system level, consideration should be given to expanding the remit of existing services to include HaH as an integral component of acute care pathways rather than as a standalone or peripheral initiative. Leveraging existing community and intermediate care structures can reduce duplication and enhance continuity, provided there is adequate investment in workforce, digital infrastructure, and governance frameworks to support this expansion. Stakeholders should be engaged early from both clinical and non-clinical backgrounds15 to support broader understanding of HaH.

There is also a need to establish dedicated HaH leadership roles, rather than assigning responsibility to individuals who already hold multiple operational portfolios. Dedicated leadership provides the necessary focus for service design, cross-sector coordination, and performance oversight, while ensuring accountability for quality and safety.21–23

Sustainability further depends on the creation of long-term financial resourcing models to underpin recruitment, training, and scaling of services. Reliance on short-term pilot funding creates instability and undermines workforce confidence. The initial funding provided by NHSE to start the HaH programmes in England, undoubtedly supported implementation, however further research is required to examine how the changing nature of the funding impacts service scale up and sustainability. Recurrent investment through long term commissioning frameworks would enable systems to build capacity, invest in skills and equipment, and evaluate impact over time.

To promote continuous improvement, systems should formalise clinically led learning from the experiences of teams that have established and evolved HaH services. Much of the current learning remains informal and localised, without drawing on long established UK services or those from countries with well-embedded services such as USA or Australia. Establishing structured, system-wide mechanisms such as professional networks, communities of practice, and shared data repositories, can enable consistent knowledge transfer and accelerate collective progress.24,25

Lastly, systems must tolerate initial variation among providers. Given differences in population needs, local infrastructure, and clinical capability, early implementations will inevitably vary in scope and design. Embracing this variation as part of the innovation cycle26 allows systems to learn from diverse approaches, progressively standardising around evidence-informed best practice rather than enforcing premature uniformity.21,27

Service-Level Recommendations

At the service level, the implementation of HaH requires realistic targets and timelines that reflect the complexity of adapting hospital-based practices to the home environment. The transformation involves not only logistical redesign but also significant cultural change across multidisciplinary teams. Allowing sufficient time for workforce adaptation and iterative learning is critical for maintaining safety and engagement.

Each service should appoint leaders with a proven record of change agency, ideally individuals with prior experience in service innovation or practice adaptation. These leaders play a pivotal role in translating vision into practice, aligning diverse stakeholders, and embedding new models of care. It is equally important that such change agents have visible and sustained support from senior management, ensuring that HaH development is recognised as a strategic organisational priority.

To maximise their effectiveness, change agents must have structured access to a range of developmental and professional activities, including:

  • Peer professional mentoring from more experienced HaH implementers both within the NHS and the global HaH community;

  • Interdisciplinary learning collaboratives, organised through professional societies or system-wide networks;

  • Leadership development programmes to strengthen strategic and operational capability;

  • Clinical skills training across the full multidisciplinary team, ensuring all staff are confident and competent in delivering care in the home environment.

These activities foster professional confidence, build shared purpose, and sustain momentum during the challenging early stages of implementation.

Finally, across both service and system levels, there needs to be greater recognition and acknowledgement that HaH services represent a complex service28 innovation and implementation and scale will take a period of time to embed before services reach their full capacity.

Limitations

This study was conducted within the Midlands region of England, but we believe many of the service level findings are generalisable across services within England. A large number of interviews with clinical staff (representing a wide range of professions), managers and policy makers were included. Whilst our study attempted to have representation from a range of geographical areas, we had more services in urban areas and future work may wish to identify more rural services. Further research could characterise the qualities that successful HaH leaders demonstrate, stakeholder perceptions at different stages of implementation and the financial impacts on implementation.

Conclusion

A range of facilitators and barriers for implementing Hospital at Home services were identified at both a service and system level. Services and systems can use the recommendations to support development and evolving service provision.

Funding

This study was funded by NHS England Midland region. This study was supported by the National Institute for Health Research (NIHR) Applied Research Collaboration (ARC) West Midlands, NIHR Community Healthcare HealthTech Research Centre (HRC) and NIHR Oxford Biomedical Research Centre (BRC). The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health and Social Care.

Data Sharing

The data for this study will not be shared, as we do not have permission from the participants or ethics approval to do so.

Conflicts of interest

None

Author Contributions

Data curation: Sophie E. McGlen (Lead), Sarah Woolley (Supporting). Formal Analysis: Sophie E. McGlen (Lead), Sarah Woolley (Equal), Graeme Currie (Supporting), David Aldulaimi (Supporting), Daniel Lasserson (Supporting). Project administration: Sophie E. McGlen (Lead), Sarah Woolley (Supporting). Writing – original draft: Sophie E. McGlen (Lead). Writing – review & editing: Sophie E. McGlen (Equal), Sarah Woolley (Equal), Graeme Currie (Equal), David Aldulaimi (Supporting), Daniel Lasserson (Equal). Investigation: Sophie E. McGlen (Lead). Conceptualization: Sarah Woolley (Equal), Graeme Currie (Equal), Daniel Lasserson (Supporting). Funding acquisition: Sarah Woolley (Equal), Graeme Currie (Equal). Methodology: Sarah Woolley (Supporting). Supervision: Sarah Woolley (Lead).

Corresponding author

Sophie E. McGlen: sophie.mcglen@wbs.ac.uk