Community hospitals and their services in the NHS: identifying transferable learning from international developments - scoping review, systematic review, country reports and case studies

The role of specialists

Given that many community hospitals described in studies included in this review do not tend to have GPs on site full time, it is perhaps not surprising that on-site presence of specialists is even less common. In most cases, the specialist tends to perform an intermittent or remote supervisory role. 59,68 Models of such supervision include weekly oversight by a consultant from the nearest acute hospital,54 or regular educational visits. 59 One study from New Zealand reported on consultant surgeons who undertake visits to community hospitals over a distance of 150 km at least twice per week. 72 In more remote areas, specialist visits may be less frequent, as in the case of the delivery of obstetrician outpatient clinics several times per year,67 or of specialist eye care offered by visiting specialists for 1 week two times a year in Australia. 44

This limited or remote specialist involvement means that GPs and nurses are required to be flexible in their roles and to demonstrate a broad spectrum of skills. 21,41 For example, GPs may perform minor surgery or caesarean sections68 and have ‘multiple roles’, which include ward duty, GP clinics and emergency unit on-call, such as in a 20-bed rural hospital in Victoria, Australia. 77 Small regional hospitals in the Northern Territory in Australia are staffed with GPs trained to perform emergency and elective surgery. 47 As for nurses, they may have to demonstrate skills in areas such as clinical procedures, diagnosis, leadership, patient-centred care, interprofessional communication, spiritual guidance and bereavement support,78 or to master some relatively complex diagnostic tools (e.g. for stroke54 or chest pain48).

The role of nurses

The importance of the nurse’s role was particularly emphasised in community hospitals where, in addition to requirements for a broader skill set, they hold greater managerial 59 and patient-related responsibility than in larger hospitals. 79 Senior nurses or midwifes are often in charge of managing a unit or the whole hospital, as in the case of the 18 community hospitals reported on in Ireland. 59 They may be responsible for the patient from admission to discharge,80 without the patient seeing a doctor. 77 In other cases, nurses were in charge of the development and implementation of a specific specialist service, such as a chronic musculoskeletal pain service42 or a mental health liaison service. 81 Steers et al. ,82 based on a review of the evidence of providing palliative care in community hospitals in the UK, concluded that GPs generally acknowledged their dependence on nursing staff to support them to make timely management decisions following the admission of patients.

Post-acute care

Four studies in seven papers evaluated health outcomes in a post-acute setting. 7,9,10,61,62,98,100 In their high-quality RCT, Young et al. 10 used the Nottingham Extended Activities of Daily Living Scale to assess independence 6 months after treatment in patients randomised to receive either intermediate care at a community hospital or usual care at an acute hospital. The study showed a significant increase in independence at 6 months in the community hospital compared with usual care [mean difference = 3.27, 95% confidence interval (CI) = 0.26 to 6.28; p = 0.03], supporting findings from the previous single-centre study by the same group (mean difference = 5.30, CI = 0.64 to 9.96). 7 In the multicentre trial, similar proportions of both groups died before follow-up (p = 0.33), and, among those living at home before admission, similar proportions of both groups were discharged to a care home, died before discharge (p = 0.08) or continued to live at home (p = 0.426) at 6 months. 10 Conversely, anxiety levels 1 week post discharge showed a greater increase in the community hospital group (p = 0.03) as assessed using the Hospital Anxiety and Depression Scale (HADS).

Additional analyses of both RCTs found a trend towards worse independence outcomes for patients who were transferred to the community hospital late or not at all (p = 0.072 for the multicentre trial). 98,100 Pairwise comparisons in both studies showed significant differences in favour of patients who were transferred to the community hospitals early compared with those who remained at the acute hospital for longer.

A high-quality RCT set in Norway compared provision of intermediate care at a community hospital with continued care at an acute hospital. 9 Care at the community hospital was associated with a decrease in readmissions for the same complaint within 60 days [odds ratio (OR) 2.77, 95% CI = 1.18 to 6.49; p = 0.03] and an increase in independence from home care (OR 1.21, 95% CI = 0.59 to 2.52; p = 0.02) compared with the acute hospital. There was no significant difference in the proportion of deaths at 6 months (p = 0.23), but at 12 months a significantly smaller proportion of patients had died in the community hospital group (p = 0.03). 62 The number of patients living in nursing homes and total number of bed-days at 6 months did not differ significantly between settings, but the acute hospital group had fewer days of initial inpatient care (p = 0.00) and the community hospital group had fewer inpatient days for readmissions (p = 0.04). 9 No significant differences were found at 12 months in the number of admissions, acute hospital bed-days or for long-term primary-level care needs, which the authors link to a loss of power to reveal differences owing to the high proportion of deaths during follow-up. 62

Chemotherapy provision

A high-quality crossover RCT compared chemotherapy in outreach locations at four community hospitals with chemotherapy delivered in a dedicated cancer centre. 66 The community hospitals had not previously delivered chemotherapy and in all settings treatment was delivered by a hospital chemotherapy team. No safety issues or adverse reactions were reported in community hospitals and there were no significant differences between the treatment groups either in toxicity levels (using Chemotherapy Symptom Assessment Scale) or in anxiety and depression (using HADS).

Palliative care

One qualitative study on end-of-life care (in two medium-quality papers) asked patients and their family carers about care in community hospitals compared with acute hospital care, and suggested that the community hospital was perceived as preferable. 45,58 Hawker et al. 45 reported that bereaved carers expressed more satisfaction with community hospitals than with their nearest acute hospital, valuing ease of access for visits, pleasant environment and facilities, familiarity between staff and families and sensitive and kind nursing staff. Participants also commented that staff seemed to care more for the patients and had more time to spend with them than at the acute hospital. These findings were supported in Payne et al. ,58 as patients reported the community hospital to be cleaner, more comfortable, better located and more flexible in responding to individual needs. Study participants also particularly valued the continuity of care offered by GP-led community hospitals. Both papers, however, reported concerns related to a lack of contact with qualified nursing staff and noise levels in community hospitals. Older patients also highlighted the lack of facilities and qualified staff for complex medical procedures, when compared with the acute setting. 58

Chemotherapy provision

In Pace et al. ’s66 crossover RCT, some 97% of patients chose to receive their remaining chemotherapy treatment at a community hospital rather than a cancer centre. Data from the Chemotherapy Patient Satisfaction Questionnaire showed that patients were significantly more satisfied with the community hospital than with the cancer centre for accessibility (mean difference = 1.05, p < 0.001), environment (mean difference = 2.79, p < 0.001), technical aspects of nursing care (mean difference = 0.58, p < 0.01) and interpersonal aspects of nursing care (mean difference = 0.4, p < 0.05) . There was no significant difference in global satisfaction between the settings.

Post-acute care

Three studies set in England reported high patient satisfaction and preference for community hospitals for post-acute care and rehabilitation across five papers. 7,8,10,54,55 One low-quality qualitative study reported that patients discharged to the community hospital for rehabilitation after stroke appreciated the convenient location, which made it easier for relatives to visit. 54 The community hospital was also perceived to be less confusing and safer than the acute hospital because of its small size, and patients appreciated having a choice of location.

Young et al. ’s10 multicentre and Green et al. ’s7 single-centre RCTs found only minor differences in patient and carer satisfaction between community and acute hospital groups. In the Young et al. 10 trial, patient satisfaction with their own recovery was similar for both groups but greater at 1 week after discharge for community hospital groups (OR 2.12, 95% CI = 1.3 to 3.46; p = 0.004). Green et al. 7 reported that patient and carer satisfaction with community services was also similar for both groups, but greater for the community hospital group at 3 months after recruitment (OR 3.43, 95% CI = 1.05 to 11.24). They found no difference between groups for emotional distress in carers, or in effect on the carer burden.

Two medium-quality qualitative papers nested within the two RCTs found patients’ and carers’ experiences of care in community hospitals generally compared favourably to equivalent care in acute hospitals. 8,55 Patients were reported to value the location, food, staff attitudes, the physical environment and the ambience of the community hospital, including single-bed accommodation, which was perceived to be more home-like. 55 These findings were supported by Small et al. ,8 who reported that patients and carers described the community hospital as more ‘homely’ and valued its accessibility. There was a perception, as expressed by one carer, that the community hospital’s home-like environment would promote recovery through encouraging the self-care activities needed after discharge. This was contrasted with views on the acute hospital, where the philosophy of care was seen to be focused on ‘medical efficiency’.

Post-acute care

Staff experiences of delivering post-acute care to older patients in community hospitals were generally described to be positive in two studies reported in two qualitative papers set in England and Norway. 8,61 In England, Small et al. 8 carried out interviews with community hospital staff, who reported that the community hospital provided a distinctly homely, calm and pleasant setting, where social interaction was encouraged and patients’ relatives were involved in patient care. Community hospitals’ orientation towards older people was also considered important. Conversely, in describing their workplace, acute hospital staff emphasised care attributes such as ‘medical efficiency’ and high standards and described a lack of stimulation for patients. At the same time, staff in both settings highlighted shared understandings of the importance of providing holistic and patient-centred post-acute care.

The study in Norway reported that the introduction of a community hospital as a ‘bridge’ between acute and primary care had improved staff experience of discharge processes, including timeliness, patient preparation for transition and communication between the different levels of care. 61 Primary care staff were satisfied that patients with complex needs were ‘shielded’ from fast discharge from the acute hospital. Staff interviewed for the study felt that, overall, the community hospital liaised effectively with both acute and primary care, with staff at the acute hospital describing the community hospital ‘as an extension of a hospital department’, whereas those in primary care commented that the community hospital served ‘as a buffer that provided preparations for discharge of the patients’.

Treatment costs in post-acute care

Garåsen et al. ’s62 medium-quality cost analysis of intermediate care in Norway was nested within a RCT and calculated average costs per day of treatment from randomisation up to 12 months after discharge from a community hospital or an acute hospital. It found treatment costs in the community hospital group to be lower than in the acute hospital group. Lower costs were mainly attributable to reductions in the number of readmissions and days readmitted, as well as lower average costs per day at the community hospital. Total average costs per patient per day in the community hospital group were £56.12 (p = 0.026) compared with the acute hospital group at £74.27 (p = 0.026) [converted from Norwegian kroner (NOK) by the authors using historical rates (GBP 0.0926 per NOK)]. Similarly, the total average cost per patient per treatment episode, including readmission within 60 days after discharge, were lower at the community hospital than at the acute hospital, at £3671.59 (p = 0.002) compared with £6798.41 (p = 0.002). There were no significant differences between groups in hospital costs during the follow-up year or for after treatment. The costs of long-term hospitalisation were highest in the community hospital group, but the difference was not statistically significant.

In two high-quality papers embedded in linked RCTs, O’Reilly et al. 11,99 reported on the cost-effectiveness of post-acute care for older people in community hospitals in England, which they found to be similar to that provided in acute hospitals. This was attributed to similar mean length of stay in hospital, similar resource use per patient and similar mean costs per patient of health and social care over 6 months of follow-up. There were non-significant differences for changes in quality-adjusted life-years (QALY) at 6 months after baseline in favour of the community hospital groups. The incremental cost-effectiveness ratio estimate was £16,324 per QALY for the multicentre study and suggested similar cost-effectiveness in the single-centre study. A cost-effectiveness acceptability curve for the multicentre study suggested that if decision-makers’ willingness to pay per QALY was £10,000, community hospital care would be cost-effective in 47% of cases. At a willingness-to-pay threshold of £30,000, this increased to 50%. 11 For the single-centre study, the acceptability curve suggested cost-effectiveness at the same thresholds in 51% and 53% of cases, respectively. 99

Treatment and drug costs in emergency care

Stapley et al. ,102 in a cohort study of uncertain quality, compared the use and cost of medicines and investigations for emergency admissions of older patients to five community hospitals and an acute hospital in England. They found the mean values for the total cost of investigations during the hospital stay per patient to be higher in the acute hospital, at £323.94 [standard deviation (SD) £245.73] than in the community hospitals, at £65.94 (SD £74.27). Similarly, the mean cost of drugs prescribed during hospital stay per patient was higher in the acute hospital (£102.74 vs. £41.86), as were the mean costs of prescriptions at admission and discharge. Overall, the median total number and cost of drugs prescribed per patient in the community hospitals was lower (p = 0.001). The authors noted that, overall, cost differences were relatively small compared with capital and staffing costs of the hospitals. The paper concludes that considerable extra resources were used in emergency treatment of older patients at an acute hospital, without yielding mortality or morbidity benefits (conclusions around clinical benefits refer to a linked paper not included in our review). 103

Patient and service costs for chemotherapy

Pace et al. 66 compared patient and service costs incurred providing chemotherapy in community hospitals and a cancer centre. Patient travel costs to attend chemotherapy in the community hospitals were found to be lower by almost half, as patients tended to live closer to the community hospitals (estimated additional patient cost of £6.29 per cancer centre visit over community hospital visit). However, when considering service costs, the need to relocate nurses to deliver chemotherapy at the community hospitals was associated with additional travel costs, both direct (£12.83 per clinic attended) and indirect (opportunity cost; £32.08 per clinic attended). Overall, the cost of delivering chemotherapy in the community hospitals was found to be marginally higher than in the cancer centre.

Data collection template and draft country reports

The data collection template sought to capture information on the basic features of the system context within which health care is governed, financed, organised and delivered more broadly. In particular, the template sought to capture the different types of models of community hospital, including the nature and scope of services provided, ownership, the interface with other forms of care, evidence of effectiveness and future development. Data were collected by means of a review of the published and grey literature, which was identified through an iterative search of the biomedical database PubMed, the world wide web using common search engines [Google Scholar (Google Inc., Mountain View, CA, USA)], and of governmental and non-governmental agencies and organisations with a remit in the area of health service and system organisation, financing and governance in the countries under review. We included documents written in English, Finnish, Italian and Norwegian. The detailed template for data collection is presented in Appendix 3.

The data compiled using the template formed the basis for draft country reports, which were shared with one expert in each of the five countries. Country experts were identified from the Health Systems and Policy Monitor hosted by the European Observatory on Health Systems and Policies112 and the authors’ own professional networks as established through earlier international comparative work. 113 Experts were asked to review the draft report for their country and to comment on and verify the information presented, and, where appropriate and relevant, to provide additional information, in particular on areas that are not well documented or that require in-depth understanding of the country context.

Key informant interviews

Interviews with key informants in the five countries sought to enhance understanding of the more salient issues pertaining to the context and processes within which community hospitals operate and function in the different systems. Key informants were identified through a combination of purposive and ‘snowball’ strategies using the published literature, official websites, the authors’ professional networks and recommendations from other study participants. We focused on a range of stakeholders involved in the organisation, governance or delivery of health services, considering representatives from national and local authorities, professional and hospital associations, and research organisations.

Potential study participants were invited to participate by e-mail and provided with background information. Interviews explored broad topics around community hospitals, including their role in the health system, types of community hospitals, funding, perceived benefits and challenges of community hospital provision and anticipated future developments, according to an interview topic guide. The interview topic guide, which was shared with participants before the interview, is presented in Appendix 4. Interviews followed ethics principles of conducting research involving human subjects. Key informants were approached in their professional role only, and no sensitive personal information was collected. As part of the consent process, study participants were assured that neither they nor their organisations would be identified in any reports of the study findings. In line with the approved study protocol, consent for participation in the interview and for the interview to be audio-recorded was sought at the beginning of each interview.

All but one interview were carried out by telephone between June and November 2015. All interviews were conducted in English. Two researchers were present during interviews, with one leading the discussion while the second person listened for key areas to explore further. One interview was undertaken in person and with only one researcher present. Interviews lasted 45–60 minutes, were audio-recorded following consent and were transcribed verbatim. Two study participants were unable to participate in interviews but provided written information according to the interview topic guide. Interview data were analysed by country according to the pre-defined topics that guided the collection of evidence for country reports (see Appendix 3). The main purpose of key informant interviews was to gather additional insight into the role of community hospitals in a given health system and their positioning in wider context of service delivery. As such, interview data were not subjected to a formal framework analysis although data were analysed following the principal steps of thematic analysis, including familiarisation of data through reading and rereading transcripts and organising data according to pre-defined topics. 114,115 Interview data were incorporated into country draft reports, which were then shared again with country experts for a final review. Country reports were updated as required to reflect changes in the policy context that concerned, directly or indirectly, community hospitals. Completed country reports formed the basis for the analysis presented here; the full country reports can be found in Appendix 5. We principally followed a thematic approach, synthesising and contrasting data according to the core themes leading the individual country reviews. We therefore begin by briefly setting out the key characteristics of the health systems in the five countries as they relate to governance, financing and principles of service organisation, in particular of hospital services. We then assess how community hospitals have evolved and we describe the types of services provided, including their role in the wider system of service delivery. We consider the evidence of benefits and the challenges faced by community hospitals in each country, and conclude with an assessment of anticipated future changes. Data derived from key informant interviews are indicated as key informant (KI) in combination with the initial of the country represented (Table 6) (e.g. KI-F03 signifies key informant number 3 from Finland). Data reported here represent reviewed countries’ policy context as of March 2016.

Defining the community hospital

Table 9 provides a summary overview of selected principal features of service delivery models in the five countries reviewed, which can be broadly subsumed under the term ‘community hospital’. Models can be seen to lie on a continuum. At one end of this continuum we observe the ‘traditional,’ small community or cottage hospital based in a typically rural or remote setting and which serves a ‘geographic purpose’ (KI-A01), that is, it provides a range of services to the local population including emergency care, minor surgery, low-acuity acute general medical care, obstetric care and basic diagnostics. These types of models are found mainly in Australia, Norway and Scotland. At the other end lie structures that are dedicated to specific population groups, primarily older people at different stages of dependency. These services may include sub- and post-acute care, intermediate care, rehabilitation, and, in some instances, palliative or end-of-life care and long-term or residential care. These structures are found predominantly in Finland, in the form of health centre inpatient wards (also referred to as health centre hospitals), and in Italy.

However, this distinction is not clear-cut. For example, community hospitals in Scotland may also provide intermediate care, rehabilitation services and end-of-life care,130 whereas community hospitals in some Italian regions, such as the Marche region, are more akin to the conceptualisation of a traditional cottage hospital which sees the provision of a wider range of services including selected medical specialist services (see Table 9). In other parts of Italy, such as Friuli Venezia Giulia, regional legislation has defined the community hospital as an intermediate care structure131 but it is mostly conceptualised in the context of residential care, as part of rehabilitation and long-term care (KI-I03).

Evolution of community hospitals

The diversity of community hospitals illustrated in Table 9 has to be interpreted in the context of the evolution of the overall hospital service delivery structure in place in different countries. Thus, some regions in Norway, such as the northernmost county of Finnmark, have a long history of operating cottage hospitals (in Finnmark this history dates back to the 1850s), which are seen to have formed a core component of health service delivery. 132 Formally defined in the 1969 Hospital Act,133 cottage hospitals in Norway are generally seen as organisational structures that deliver health care in rural areas, have overnight inpatient beds and typically also provide a selection of specialist outpatient services (KI-N04). This description also includes small general hospitals that have experienced changes in the service delivery structure over time in response to changing local population needs. Examples include the Hallingdal hospital in southern Norway (see Chapter 2), which, from the 1980s, has gradually developed specialist health-care services in an attempt to reduce acute hospital admissions. 21 More recently, such localised efforts have been placed on a more strategic footing with the 2012 Coordination Reform, which sought to enhance integrated care in the community and improve co-ordination between different levels of care as part of an overall movement to shift services from the hospital sector into primary care. The reform introduced new structures in the form of municipal acute care beds [kommunale akutte døgnplasser (KAD)] to complement or potentially substitute the traditional cottage hospital model. 108 KAD can be set up in a variety of settings [e.g. nursing homes, (newly established) local medical centres, wards in a hospital]. They aim to address the needs of patients with acute but manageable conditions, who need 24-hour supervision, but not necessarily in an acute hospital setting. All municipalities in Norway are required to provide municipal acute care beds from 2016. 134

Similar to Norway, many community hospitals in Scotland pre-date the formation of the NHS in 1948. 13 Often originally established by local benefactors in response to the needs of the local community, many have subsequently evolved into what has generally been described as GP hospitals. For example, by 1980, some 70 hospitals in Scotland could be classified as GP hospitals, estimated to account for some 3% of all hospital beds across the country. 135 Despite forming an integral component of the service delivery system in Scotland, in 2006 community hospitals received renewed attention as part of a wider government strategy that foresaw an expansion of community-based, primary care-led services. 31 The strategy envisaged new models of community hospital to operate as ‘local community resource centres’ for the provision of ‘more holistic and integrated services quicker and closer to home’. This was followed by the 2012 Community Hospitals Strategy Refresh, which sought to place the 2006 vision for community hospitals in the context of more recent strategic priorities, seeing them as having a key part to play in the health and social care integration reforms, acting as a base for community services, or as a community hub for health and social care services. 13 The role of community hospitals was also seen to have been strengthened in the context of the 2020 strategic vision for the delivery of health-care services in Scotland, with its aims to provide care close to people’s homes within an integrated health and social care system. 136 Compared with Norway and Scotland, the evolution of community hospitals in Finland has been a fairly recent phenomenon, following the 1972 primary care reform which established municipal health centres. These emerged in part from small local or district hospitals and as a result they offer a wide range of services, including GP-led inpatient units (also referred to as health centre hospitals). 117 Newly formed municipal health centres that were not linked to a former local hospital also established inpatient wards. However, in contrast to the traditional community hospital in Scotland or the cottage hospital in Norway, health centre inpatient wards aim primarily to address the needs of dependent older people, who do not require hospitalisation but who cannot be cared for in their own homes (KI-F04). This focus on older people has intensified over time, following a reform in 1993 which made municipalities responsible for all secondary and tertiary care costs. 117 It led to a change in the composition of patient population cared for in health centre inpatient wards, as ‘they lost much of their professional attraction and gradually became long-term care facilities for the elderly, especially those with memory disorders’ (KI-F05). 117

Somewhat similar to Finland, community hospitals in Italy also evolved more recently. They were set up from the mid-1990s onwards, partly in response to the changing needs of an ageing population. 137 They seek to provide medium-to-low-level clinical management of chronic degenerative diseases and of predominantly older people with a range of social care needs. 138 The development of community hospitals in Italy can also be seen as a response to national policies that from the 1990s onwards have sought to make acute care provision the primary focus of hospitals and to reduce the length of acute hospital stay. Community hospitals are expected to provide a ‘protected place’ for patients who are not yet sufficiently stable to be cared for in their own homes following discharge from acute care (KI-I01, KI-I02). They also provide a solution to the threat of hospital closures through the repurposing of existing local hospitals into multifunction facilities:

So we save the small local hospital, we keep it over there. We try to name it hospital . . . its written in big [letters] . . . so they [patients] feel comfortable to go there, but I mean no expert would ever name it a hospital.

KI-I01

Importantly, the development of community hospitals in Italy has largely relied on regional efforts. Indeed, it was only in 2006 that community hospitals were formally mentioned in a national-level document in the context of a proposed reorganisation of primary care, highlighting the need for the development of intermediate care services and of GP-led community hospitals. 139 The potential value of this form of service delivery was formally recognised following the 2012 legislation on the reorganisation of hospital care, with the 2014–16 health agreement mandating the establishment of community hospitals for the delivery of subacute care, seeking to reduce the number of inappropriate admissions to acute care hospitals and to facilitate the discharge of patients from acute care. 140

Although the notion of a community hospital is difficult to clearly delineate in all countries reviewed above, it remains a particular challenge in Australia. This is in part because of the way health services are governed in the Australian context, with states assuming responsibility for regulating hospitals. Hospitals tend to be small, particularly in regions that cover large geographical areas. The majority of beds tend to be in larger hospitals and in more densely populated areas, whereas some 70% of hospitals have 50 or fewer beds. 141 This poses considerable challenges, with an observed increase in closures of a number of small hospitals over the past decade:142

Rationally, they can’t operate [. . .] and the occupancy rates is so low, it’s very hard to argue that they’re operating efficiently and it’s increasingly evident that it’s hard to operate things safely, for anything with fairly low security levels.

KI-A05

At the same time, such facilities are seen to be indispensable in remote communities ‘where a patient can be stabilised and evacuated because the distances are significant’ (KI-A01), although this service might not necessarily be provided by a hospital as such. In response, most states have started transforming small rural hospitals into alternative settings, such as residential aged care facilities, community medical centres or multipurpose health centres. For example, states have invested in the redevelopment of small rural hospitals into so-called ‘multipurpose services’ as part of the Multipurpose Service (MPS) Program, which was established in 1993. 143 A joint initiative of the federal and state governments, these services are designed specifically to address the needs of small rural and remote communities that could not viably support stand-alone hospitals or aged care homes. MPSs may or may not include inpatient beds and are not generally considered ‘community hospitals’ in the Australian context.

Nature and scope of services delivered

Although countries, or regions within countries, offer a wide range of community hospital and similar service delivery structures, one area of commonality is that the local GP usually holds clinical responsibility and has oversight of the medical care provided, as well as admission rights. This reflects the general function of community hospitals, sitting ‘in the middle of the patient journey between home and the district general hospital’ (KI-S03). In some instances, having a GP lead can be seen to have evolved out of necessity, particularly in the cases of rural and remote regions, due to the fact that ‘the specialists aren’t there’ (KI-A01). Elsewhere, the central role of the GP has been a consequence of the historical development, such as in health centre hospitals in Finland, cottage hospitals in Norway and community hospitals in Scotland, as noted earlier.

Where community hospital-type delivery models are newly established, such as in Italy and Norway (municipal acute care beds), the formal assignment of clinical authority to a GP has been intentional because these structures are seen to form an integral part of the (reorganised) primary care system. 134,140 However, in some instances, and reflecting the specific function of the community hospital, the clinical lead may also be a geriatrician. This is the case for some structures in the Veneto region in Italy144 and for geriatric hospitals in Scotland, which have primarily long-stay beds, although there is some debate over whether or not this latter example should be included within the community hospital definition (KI-S03, KI-S05).

Service delivery typically involves multiprofessional teams, which, in addition to GPs, may comprise nurses, health-care assistants, and allied health professionals (most commonly physiotherapists and occupational therapists). Health centre hospitals in Finland (KI-F02) and community hospitals in some regions in Italy may also include social workers as part of their teams,145 and community hospitals in Scotland have seen an increase in the co-location of social care practitioners in recent years. 13 In terms of the actual management of the hospital, in all the settings reviewed, this is typically the responsibility of a head nurse or nurse manager.

The involvement of medical specialists generally varies according to the range of services provided, the size and location of the community hospital, and the population serviced. For example, in Finland, health centres in larger municipalities would typically have medical specialists in internal medicine and geriatrics, for example, available on site. Community hospitals in Italy have to ensure access to specialist advice on site, which is frequently done through co-location of community hospitals with a so-called ‘house of health’, which has been established in a number of regions. Houses of health oversee access to and provision of health and social care services in the district and their aim is to provide a single point of access and initial needs assessment through the co-ordination of primary care units with other service providers in the district, including specialist and inpatient care, public health, and mental health. 146 As a consequence, community hospitals in Italy might have access to a wide range of specialist clinics.

Elsewhere, community hospitals tend to rely on visiting specialists from the acute care hospital. For example, in Scotland, patient care at the community hospital is typically provided by multidisciplinary primary care teams, with support from secondary care specialists as required. 13 Similar models are in place in Australia, where visiting clinicians provide specialist expertise, although the precise mechanism by which this is facilitated varies across states. For example, in the Northern Territory, local GPs would typically have a network of specialists whom they can phone for advice or for evacuating patients if required. In these cases, ‘the specialists fly in and out, and, you know, run clinics or have surgical lists, that sort of thing’ (KI-A02). In Queensland, small public hospitals are managed by one of the 17 Hospital and Health Services and the community hospital would be embedded within wider service provision, which might be facilitated by telehealth links between the main hospital and the local small hospital (KI-A01). Such an approach can also be found in Norway; one example is the aforementioned Hallingdal hospital in Norway, a small hospital with 14 intermediate care beds, which is managed and funded by a general hospital (see Chapter 2). 22 The inpatient department is led by GPs who are supervised, via telephone, by specialists located at the general hospital; outpatient clinics are also hosted by specialists visiting from the general hospital.

Attracting health-care workers and skills

Key informants from across the five reviewed countries all noted workforce challenges in relation to the delivery of care in community hospitals. One particular issue noted by commentators from Australia, Finland and Scotland was the difficulty to attract health-care workers, especially in rural or remote areas. However, this challenge has to be seen in the general context of workforce issues in the different settings; for example, key informants from Finland highlighted the problem of recruiting a sufficient number of doctors to work in municipal health centres more generally (KI-F05):

[T]o work on the [inpatient] ward it’s more demanding than in [outpatients], so that it is not so easy but because if you go to work to the health centre you are [a] worker there then the employer can decide that you are working in the open or in the ward. But of course, we always try to find people who really can and want to do those works because there are a lot of jobs to choose [from] so that if they say they have to go there then the workers say goodbye.

KI-F02

Key informants from Scotland also noted that an overall shortage of GPs and nurses in the country was further compounded by the fact that priority was typically given to staffing acute hospitals, which means that community hospitals can become short staffed (KI-S05, KI-S04).

There was agreement among key informants from Australia, Finland, Norway and Scotland that working in a community hospital required some additional skills, for example the ability to treat complex and acute cases (KI-N01). Commentators from Scotland highlighted that it would take some time for new staff who have received standard nursing or GP training to develop the additional skills required for working in a community hospital (KI-S05).

In Australia, GPs working in small rural hospitals or municipal health centres traditionally had a wider range of skills, meaning that they were able to perform, for example, basic surgery following some additional training. This model of ‘procedural GPs’ was reported to have been common (KI-A02, KI-A03) but to have ‘waned over a period of time, as there were closures of small country hospitals across the country because of cost pressures’ (KI-A02). However, it was noted that the general policy direction in Australia now was to (re-)encourage the development of rural generalists, that is, GPs who have additional skills, such as in anaesthesia or surgery or obstetrics (KI-A01, KI-A02). Training is provided through the Australian College of Rural and Remote Medicine, for example. 152 In Scotland, there is a recognised need to provide further support to GPs in particular, with investment in additional skills training such as the Rural Fellowships in General Practice. 153 In Norway, workforce issues were recently acknowledged by the government, which proposed to rethink hospital staffing and medical training in the context of the recent health services changes. 154

However, challenges remain, in particular where national workforce policies conflict with local requirements. For example, in Australia it was noted that while the number of medical graduates has increased, this increase was unlikely to address the challenge of attracting them to work in rural and remote settings. This was in part because GPs can set up practice anywhere in the country, and most settle in inner cities, and ‘there is nothing you can do to reallocate those GPs, you have to attract them to small rural communities’ (KI-A03). A similar challenge was reported in Finland, where one potential solution has been to provide opportunities for GPs and nurses to work with specialists in, for example, internal medicine or geriatrics (KI-F02). This was seen to be particularly attractive for more junior health-care workers given the characteristics of the patient population, that is, older patients with complex health and care problems. One key informant highlighted the need to offer flexible terms of employment and to allow workers to for example ‘continue their studies or do research, or work in private sector also [. . .] It has to be some kind of a combination that this kind of a people and specialists are willing to work in a public sector and city hospital’ (KI-F03).

A challenge of a different nature was reported for Italy, where policies introduced by the national government in the wake of the 2008 financial crisis involved a cut in workforce capacity among health-care workers, and a concurrent fall in national health-care expenditure of about 5% between 2011 and 2014. 155 Much of the cuts in the physician workforce have been achieved through a freeze on the restaffing of vacant posts following retirement. This has created challenges, as the restrictions on replacing staff coincides with an increasing need for services, and, regardless of cuts, medical staff have ‘to continue with the job, because if you have beds and you need to ensure 24 hour [care] you need to have at least six nurses, six nurses in a day, because you need to [cover services 24/7]’ (KI-I03).

The role of the community hospital within the local service delivery structure

There was a perception among key informants from several countries that as delivery systems are evolving, boundaries between services provided by community hospitals and those offered elsewhere locally might blur or indeed, overlap, impacting on the effectiveness and efficiency of service delivery, and potentially undermining the value of the community hospital and similar structures. For example, commentators from Norway expressed concerns about how the various service innovations that have been promoted following the 2012 Care Coordination reform sit alongside each other, in particular with reference to the newly introduced municipal acute care beds (KAD). There was some apprehension that the shift towards the development of intermediate care facilities at the municipality level might create tensions with local cottage hospitals and small general hospitals, which will have to compete not only with each other, but also with the newly introduced structures for funding (KI03).

Key informants in Scotland highlighted a further set of challenges arising from service restrictions elsewhere. For example, there were concerns about a perceived rise in the use of the community hospital as a step-down facility from the acute hospital (KI-S04), owing to the pressure to free bed use in acute hospitals. This was seen to have reduced the number of beds available for GP admissions (KI-S05). It was also noted that there had been delays in discharge from the community hospital to community care because of a lack of care home placements or the availability of home carers, which in turn would lead to a longer length of stay at the community hospital than was seen to be appropriate (KI-S05, KI-S04).

A related issue was reported by key informants in Italy, who described the challenges faced by regions seeking to balance the implementation of national level reforms, which place restrictions on the maximum number of hospital beds and length of stay in acute care, with an increasing need to deliver services that address the combined pressures arising from multiple chronic disease and population ageing (KI-I02). Against this background, the way that local or regional systems have evolved can be regarded as opportunistic rather than needs-based, driven by the need to reduce capacity in the acute care sector and which may impact on the delivery of appropriate care to those in need (KI-I-01):

So you may have in one local health authority twice as much subacute as the other one, the neighbour local health authority, who has half of these and twice of those. Or three times nursing beds or four times this here, so very random. So, what we have found is that, the sum of all this kind of beds are moving towards an increasing number because of the reduction of [acute] hospital beds and because the development of nursing homes, paid by families. But the combination is quite emergent, with no real rationale [. . .] But having such big differences in the mix, in the infrastructure mix, what happens is that, independent of your physical or psychological condition, you can end in one of those settings just depending on the offer you find over there.

KI-I01

A perhaps extreme example of an opportunistic approach seeking to reconcile national stipulations with local need is that taken by the Marche region in Italy, which, in response to the nationally mandated requirement to establish community hospitals, proposed, in 2015, to redefine its 13 houses of health as community hospitals. 156

Pursuing a needs-based approach may be particularly challenging in places where community hospital-type structures have developed historically, which can result in discrepancies between, or duplication of, services delivered in the same districts. We have already highlighted concerns expressed by key informants in Norway as they relate to the implementation of new services models vis-à-vis existing ones. It was also reported to pose challenges in Finland, where some larger municipalities operate city hospitals, in addition to health centre inpatient wards; these should be complementary to the specialist services provided for by the hospital district (central and university hospitals) but services frequently overlap (KI-F01, KI-F04). In Scotland, extended role community hospitals provide a larger range of services than the typical community hospital (such as more complex types of surgery), as these have commonly developed from previous district general hospitals, and they maintain larger sites. These hospitals have often assumed a community hospital function following the development of a new acute hospital nearby (KI-S03), which poses the risk of service duplication.

Stakeholder interviews

Semistructured interviews were conducted at each site (see Table 10). Potential participants were identified via a stakeholder mapping exercise and were sampled purposively. Most interviews were conducted during field visits, with a small number of telephone interviews conducted shortly before or after. Interviews lasted for between 30 and 60 minutes. Interviews for Case A were conducted by JC and EvT and for Case B by JC. A semistructured interview guide was used, although the approach remained flexible, both for the researchers and the participants, and participants were encouraged to talk from their own perspective. The common interview guide covered origins and likely future development of community hospitals; staffing and staff roles; local system of services; relationships with other providers and community; vision for the community hospital; and challenges to provision of services (see Appendix 7). Informed written consent was obtained from each interviewee and interviews were audio-recorded and transcribed verbatim, with permission. Permission to record was refused only once and notes were taken instead. Transcripts and notes were anonymised.

Non-participant observation

Non-participant observation was conducted in the selected community hospital settings as a means of gaining an understanding of organisational functioning. Observations were conducted in the public areas of each community hospital (reception, corridors, waiting rooms) and during multidisciplinary team meetings. No observations of patient care were conducted. Observations were conducted on a variety of days of the week and times of day in order to gain an understanding of the full range and levels of activity. Observations were recorded using a semistructured observation framework. 164 Management permission for non-participant observation was sought before commencing field visits but consent was not sought from everybody observed. Names or identifiable information was not recorded as part of observation notes.

Document review

Documents describing the organisations, their mission, status, activities and working relationships were reviewed to provide a historical perspective and contextualise the hospitals’ development and functioning. Documents were sourced through internet searching and signposting by stakeholders and included records related to the structure, infrastructure and governance; community hospitals’ setting within the wider health system; minutes of management meetings; and communication materials. In addition, we profiled the population served, using routinely collected anonymised data. 165,166

Providing both step-up and step-down care

Staff at both hospitals shared a vision of the ideal community hospital role within the wider health system. It focused primarily on the provision of timely, effective preventative and step-up care locally. Especially with regard to elderly patients and those with chronic illnesses, they sought to keep patients out of hospital where possible.

However, both community hospitals regularly accommodated patient transfers from overstretched nearby acute hospitals, providing step-down beds to ease patient discharge through intermediate care and often taking on delayed discharges to relieve pressure from the wider system. With shortages in home care causing delayed discharges in both settings, it was often challenging to balance the demand for step-down beds with the imperative to provide step-up care to local populations.

Establishing a flexible service model to meet changing needs

Despite the range of primary and specialist care services offered at the community hospitals, staff and management in both settings felt that the potential for local provision had not yet been fully realised. As local elderly frail populations grew, building acute skills and being able to expand flexibly the range of services at the community level were seen as key to the community hospital role within regional health-care systems and their relationship with acute services. The delivery of ‘enhanced’ services, towards the specialist end of the care spectrum, and the ability of GP, nursing and allied health staff to support specialist services were seen as important ways for the hospitals to relieve the pressure on acute provision and to improve patient access. In Case A, some GPs were trained in specialist procedures such as minor surgery and joint injections, as well as dermatology. Satellite services by visiting consultants were another approach to service diversification. In challenging GPs and other staff to extend their skills in this way, both hospitals served to address a wider clinical skills gap in the community for dealing with complex patients.

The ability to provide a community resource that could adapt to accommodate future needs and complement regional acute provision was crucial and as such, had been a key design consideration during the facilities’ redevelopments. In Case B, one participant described this as ‘future-proofing’ (S1_B).

Providing integrated services from a central hub

Both hospitals were designed to serve as ‘hubs’ for a wide range of health and social care services that were previously separately located (see Table 11), and to therefore catalyse improved integration and collaboration across teams and disciplines. The primary reported benefits of co-location in both settings centred around the facilitation of informal interactions and relationship-building between teams. According to participants, this allowed staff to communicate more efficiently and to understand each other’s roles.

Weekly multidisciplinary team meetings were the foundation of interservice co-operation in both settings, where discharge plans were discussed with input from representatives of relevant services, including GPs, nursing, physiotherapy, district nursing, occupational therapy, adult social care and mental health. Participants described how co-location had directly benefited patients, enabling more joined-up care, quicker referrals and closer communication between disciplines. In both hospitals, GPs and other professionals appreciated having easier access to diagnostic facilities than they had had previously.

In both settings, the integration (formal or otherwise) of health and social care services was particularly helpful in responding to home care shortages, although this occurred in different ways. In Case A, formal integration had allowed the Health Board to fast-track the registration of a new home care operator, whereas in Case B, close co-operation between the integrated community health (including hospital at home, intermediate care and district nurses) and home care teams had led to the development of joint care for patients in the community.

Providing local, holistic and patient-centred care in a home-like environment

At both sites, a key defining function identified by participants was the provision of care close to home that was personalised, holistic and patient-centred, often defined in contrast to care in larger acute hospitals. The role of GPs in the leadership and delivery of hospital services was central, as their prior patient knowledge and relationships were considered beneficial to patient experience as well as to the community-oriented character of the hospitals more broadly. The importance of community feeling and a sense of duty to the local population were also clear in the operation of the hospitals. This was particularly visible in Case A, as staff sought to prioritise admissions from the local community.

Participants reported that by facilitating more joined-up working, the integration of services at the community hospitals led to a more holistic and personalised approach to care. The hospitals’ homely and friendly environment was also appreciated by patients and staff, and the decor and layout of the wards in both hospitals counteracted the medicalised setting. Walls were observed to carry community notices, pictures and local historical information. Wards were spacious and bright, with well-used common rooms for patients, with views looking out on to the grounds. For elderly inpatients in particular, the community hospitals were considered better for recovery than acute hospitals.

Balancing the optimum community hospital function with wider system needs

Interviews revealed a tension in both community hospitals between their optimum role and the needs of resource-constrained systems, in particular in relation to long-term care of delayed discharges and bed shortages in nearby acute hospitals. Participants in Case A in particular emphasised a tension between their own understanding of the hospital’s function as a provider of step-up care, and expectations of wider health system actors. However, although Case A regularly accommodated acute referrals, local admissions were prioritised and efforts made to retain a ‘community bed’ at the weekend, just in case a local patient needed it.

This sense of competing objectives was potentially reinforced by blurred boundaries between the levels of care and misperceptions of the function and limits of community hospital provision. In both cases, staff reported that acute hospital colleagues had limited understanding of the community hospitals’ capabilities: ‘I think our understanding of what goes on in hospitals is possibly better than the hospitals understanding of what we can do in the community’ (S1_B). In Case A, a staff member reported that acute hospital clinicians were at times reluctant to refer certain patients to the community hospital as they underestimated staff skills. Conversely, in Case B, it was reported that pressure to free up beds at times led to the premature transfer of acute patients with care needs that stretched staff’s skill sets.

In addition, in seeking to expand the delivery of specialist services, the hospitals also faced the challenge of managing the ‘tension between local and central around efficiencies and costs’ (S4_B) and the tendency for long-term gains around patient access and local skills to be stymied by expense and efficiency concerns. In Case B in particular, under-utilised facilities, including a fully equipped minor surgery suite, highlighted the complexity of planning for future community service needs.

Enabling general practitioner ownership and continuity of care

The integration of GP services into the community hospitals was deemed beneficial in two key ways: (1) they provided valuable local input and leadership to wards; and (2) the traditional GP role was enhanced by improved access to facilities and treatment options (e.g. hospital at home, diagnostic facilities, beds). The degree to which they were able to effectively and confidently fulfil their roles on the wards and in the multidisciplinary hospital environment more widely was affected by both the system used among GPs to share responsibility for ward cover, and how their role was funded and defined by the Health Board.

In Case A, GP leadership was considered key both to driving the redevelopment of the hospital and to maintaining services. The rota system used allocated one GP to ward cover for 1 full week at a time, and minimising GP rotation was considered important to the maintenance of skills and continuity.

In contrast, in Case B continuity of care on the wards and maintaining GP skills and confidence had been a challenge. Inefficiencies were created by the division of responsibility and inpatient load among three separate practices. One GP from each practice took primary responsibility for ward cover and thus maintained a regular presence but, in their absence, some of the other GPs struggled. This occasional lack of continuity, combined with a heavy workload and time pressures, reportedly made it difficult for some GPs to take confident ownership of patient care on the wards:

They know there’s no way they can get a complete handle on the patient, in one sitting, without sitting and reading the notes, and spending half an hour talking to the nurses about what’s going on.

S8_B

Contractual and funding arrangements defining the GP’s support for the hospitals, beyond their practice responsibilities, were also key to the GPs’ function. Whereas in Case A, GP input was contracted through a clearly defined service level agreement, which could be supplemented when enhanced services were added, in Case B, the complexity of GPs’ roles were reported to be insufficiently defined in contractual and funding arrangements with the Health Board. A GP reported that, although the move to the new facility had significantly increased the complexity and volume of workload, GPs still received the blanket funding for bed cover that they had previously. One participant described the mismatch:

You’ve got multidisciplinary team meetings, there is that complexity, you need to speak to the [occupational therapist] OT, the physio. So it’s completely changed, but it’s still funded as if you were just, you know, it was just like a home visit or something like that.

S1_B

Given their pivotal role as managers and leaders, participants in both hospitals highlighted the importance of respect for GP autonomy in order to sustain the model of service delivery and staffing.

Meeting community hospital staffing and training needs

Referring to GPs as well as other service staff, participants in both hospitals repeatedly emphasised that community hospitals attracted a ‘different’ kind of individual, who sought to stretch their skills and challenge themselves beyond their traditional role.

Management staff in both hospitals underlined the importance of encouraging innovation and exploiting the smaller setting to do things differently:

We have been encouraged and allowed to be innovative, to try different ways of working, to have that discussion and to get on with it to make a difference . . .

S10_A

However, understaffing was reported as a pressing issue in both hospitals, especially among nursing and home care services, which necessitated flexibility and close collaboration among the small teams to maintain services. Maintaining and building the skills required by the community hospital setting was an ongoing challenge for both hospitals. Staff often dealt with a wide range of techniques too infrequently to maintain expertise and confidence. Although nurses and GPs were commonly trained in more specialist procedures, in the context of increasingly complex and frail patient loads, staff expressed concern about having the necessary skills:

We sometimes feel a wee bit isolated when, you know, something . . . a patient’s really quite unwell and we can’t just run along the corridor or the consultant’s not just along the corridor to give that extra advice. They’re on the phone but, you know, they’re not just next door to come along and see a patient.

S10_B

In Case B, achieving the right balance between GP and consultant oversight of palliative care also demonstrated the challenge created by delivering more specialist care in the community hospital. Interviews indicated that some GPs lacked the confidence to take some palliative care decisions, preferring to defer them until the consultant’s weekly visit.

Overcoming challenges of co-location and integration

Although service integration and co-location were generally viewed positively, interview participants also identified challenges related to the sharing and division of workspace, differing work cultures and a lack of shared information technology (IT) systems. These concerns were reported to affect the degree to which the full potential of joined-up working was realised.

In Case A, where responsibility for adult social care was already under the Health Board, integration was seen as a long-running work in progress by the health and social care teams. According to interviews, different cultures of working and difficulties around defining team workspace and perceived differential power relationships were the main challenges. In Case B, the health and social care teams’ co-location within an open-plan office was reported to have improved collaboration, despite no formal integration having taken place. Shared break and dining areas in both facilities reinforced the co-operative environment.

A lack of upfront planning and investment in IT system interoperability limited service teams’ abilities to reap the full benefits of co-location. Although some IT systems in both hospitals did enable information sharing, separate systems were in use in both hospitals and they did not ‘speak’ to each other at all. Initiatives of limited scope demonstrated the value of interoperable systems. In Case A, GPs had shared access to their system with community staff, and found it extremely helpful to book physiotherapy appointments directly. In Case B, the integrated community health team’s shared system was also reported to have helped to build efficient and collaborative working practices.

Defining the ‘community’: equity, management and mission

Although Case B was managed by a community health partnership (CHP), which also oversaw two other community hospitals, Case A maintained and defended strong locality-level control. These differing management models had a strong impact on the approaches of the two hospitals and were reflected in their understanding of the ‘community’ they served.

With a perspective on district-wide needs, the CHP placed no formal boundaries on Case B’s community served and consciously extended services to patients beyond its immediate locality. The effect of this approach was most clearly manifest on the wards, where many beds were occupied by referrals from nearby acute hospitals and patients from the wider region. Health Board investment in the redevelopment of Case B was justified in part by this wide geographic coverage and the hospital’s ability to address district-wide needs:

. . . if you used a locality here to provide services you’re taking some of the pressure off another part of the system. So that was to our mind was the quid pro quo, you know.

S4_B

This function was also important to addressing equity concerns in service provision across the local region.

District-level planning was also seen as an effective approach to rationalising resources. Close relationships between the CHP’s three community hospitals strengthened local service delivery. For example, outpatient clinics were co-ordinated between the CHP hospitals in order to avoid overlap and if one was full, a patient could often find a bed in another.

However, the emphasis in Case A differed significantly, as staff emphasised a strong sense of community boundaries and the value of locality control for effective planning and resource management. Although structurally part of a district partnership, management was still taking place at the locality level in practice. This locality focus was also reflected in interactions between the ward and the acute hospital, where staff strongly prioritised local admissions over transfers of patients from beyond their locality and proactively reached out to other regional services to bring local patients into their care, where possible.

Community ownership and engagement

Despite the differences in approach to defining their ‘community’, in both settings a sense of community ownership, through engagement and use of facilities, was significant. Community engagement was seen as important, to the hospitals’ initial development, as well as their day-to-day operation and mission. It was also seen as a feature of the hospital models that distinguished them from acute hospitals:

The community hospital is at the centre of community so you should be reflecting on what the community is thinking.

S8_A

In both sites, volunteering was an important way for local people to engage in the life of the hospital. Local chapters of the League of Friends (Attend, London, UK) worked actively with staff to fundraise for equipment, for example. In Case B, volunteers maintained a continuous presence, directing patients and visitors in the reception areas, and visiting inpatients to bring newspapers, serve tea and keep them company. A staff member described the effect on the hospital’s function:

There is always a real community feel in that the community gets behind the community hospitals and they’ll fight for it.

S10_B

In both hospitals, staff participants were eager to leverage these aspects of the service models to develop the community hospital as a ‘community hub’, which could function as a space to deliver health messages even to those not receiving treatment as well as to bring people together and alleviate social isolation. In Case A, participants highlighted the importance of hospital facilities for social interaction, following complaints about the lack of a café and from elderly patients who had been attended to too quickly without having time to enjoy the outing. In Case B, the hospital’s café was observed to be continuously busy and according to interviews, was often used by members of the public without a hospital appointment.

Dialogue with the local community ensured buy-in at the planning stages. In Case A, although a community representative was included on the redevelopment panel, no formal public consultation took place. Despite the fact that the vocal and at times oppositional engagement of local media and community with hospital matters signalled the strong sense of ownership felt, the absence of a formal mechanism for direct engagement meant that some participants still sorely felt the lack of a community voice and channel for information-sharing. The picture differed considerably in Case B, where comprehensive public consultations at the planning stage facilitated community buy-in.

Description of methods used across the different countries

The same methodological approach was used across all four case studies to examine the role and function of the selected community hospitals in the three countries. Table 12 details the three main data collection components: stakeholder interviews (informed by a stakeholder mapping), non-participant observation and document review. Researchers followed a common protocol. For each case, site field visits were made, lasting between 4 and 10 days. Observations were carried out on different days and at a range of times in order to capture a variety of activity and any differences in workload and staffing within the community hospitals.

When interviews could not be completed during field visits, these were completed by telephone. A semistructured interview guide (see Appendix 7) and observation framework was used,183 although the approach remained flexible, both for the researchers and the participants. Informed written consent was obtained from each interviewee and interviews were audio-recorded and transcribed verbatim, with permission. Permission to record was declined once, and in this case, notes were taken instead. Researchers in Italy and Finland were bilingual and translated interviews into English. Permission for non-participant observation was sought before commencing field visits but consent was not sought from everybody observed. No names or identifiable information were recorded as part of observation notes. Data collection took place between November 2014 and August 2015.

Analysis

Each case study was analysed individually in the first instance. Detailed case descriptions were developed in order to describe the history, context and organisation of the community hospitals. Thematic analysis was conducted based on principles outlined by Boyatzis. 167 For the analysis of Case A, a coding frame was developed by JC and SB based on the interview topic guide and initial reading of interview transcripts and field notes. Using the coding frame, JC and SB each coded half of the transcripts for Case A, meeting regularly to discuss themes and refining the coding frame in an iterative manner as required. As analysis progressed, codes were organised into overarching or organising themes, using NVivo 10. Sourced documents, observations and field notes were read (and reread) to identify contextual and historical data and to triangulate interview findings. The coding frame for Case A was used for subsequent cases, with further refinements based on emerging findings. Cross-case analysis was then undertaken where case-study reports were systematically compared and contrasted according to the major themes identified previously. Context was emphasised in the cross-case analysis in order to understand similarities and differences between the cases. Three researchers (EP, EvT and NE) independently compared and contrasted the four case studies and identified emerging themes before discussing and agreeing themes as part of an iterative process of writing and analysis.

Health system and policy context

Chapter 4 provides an overview of the sources of health expenditure, governance, hospital bed capacity and number of community hospitals in the three countries. Italy and Finland are highly decentralised, with municipalities, or regions, having responsibility for the funding and organisation of health care, whereas in Scotland the responsibility is retained by central government (although devolved from the UK government to Scotland) with planning and delivery functions lying with territorial Health Boards. The number of community hospitals varies by country but is lower in Italy, relative to population.

Chapter 4 further summarises key legislation and policies in each of the countries that are relevant to community hospital provision. The policy context provides useful background against which to understand the different hospital models considered. It can be seen that in Scotland since 2011 there has been a series of strategies and policies specific to community hospitals and a bill to integrate health and social care. In contrast, in Italy driving policies are largely around reductions in hospitals, hospital beds and length of acute hospital stay, and in Finland around merging municipalities. This may in some part be a reflection of the longevity of community hospital within the three countries. In Scotland, community hospitals predate the setting up of the NHS in 1948. In Finland, community health centres, which have several of the same features as community hospitals, have been established since 1972, and in Italy community hospitals were established only in the mid-1990s. Community hospitals are thus at a different stage of evolution in each country. In Finland and Italy they may be seen as part of a reaction and solution to wider reorganisation within the health system, whereas in Scotland they have become perhaps more embedded in the health system.

Community hospital provision

‘Community hospitals’ are not necessarily well defined, even in Scotland where they are long established. Chapter 4 provides an overview of the definition and service focus of community hospitals in each of the countries. As noted, Finland does not use the term ‘community hospital’ but its health centre hospitals (or, more precisely, health centre inpatient wards) can be seen to be similar in terms of structure and function. We use the term ‘community hospital’ to encompass these health centres and equivalent models for the remainder of the chapter. As noted in Chapter 4 community hospitals tend to serve a ‘geographic purpose’ with a broad range of services, including low-acuity acute general medical care, whereas community hospitals in Finland and Italy tend to be dedicated to specific population groups, primarily older people, and have a service focus on sub- and post-acute care, intermediate care and rehabilitation, with some variation. An important similarity across all of the countries studied here is that GPs, often with additional specialist training, have a key operational role, leading the management and medical care. Community hospitals in Finland and Scotland offer a comparable number of beds (between 20 and 60 typically), whereas those in Italy tend to offer up to 20 beds, although this varies within and across regions (see Chapter 4). In Italy, this may be representative of the move from providing the majority of primary and secondary care in district and regional hospitals to community hospitals; a transition, which has largely already taken place in Scotland and Finland.

Defining the community hospital

It is difficult to define a community hospital in the Australian context. Key informants interviewed for this study suggested that the concept has to be understood in the context of the general way the hospital infrastructure is organised in the different states:

[Hospital models] are driven by not only by state government policy but also by the demography really [. . .] [For example, in the Northern Territory] the population is very scattered over very vast lots of very small communities. We really don’t have community hospitals per se [. . .] most of these very remote communities we have something that you would characterise more as a community health centre [. . .] We do have in a couple of regional centres larger hospitals that are largely GP run and they are in larger population centres of you know 3000 to 6000 that sort of size. They will have visiting specialist services and they would still have procedural GPs in one of those towns but not in the other.

KI-A02

In Australia, hospitals tend to be small, particularly in regions that cover large geographical areas. The majority of beds tend to be in larger hospitals and in more densely populated areas. In 2013–14, some 70% of hospitals had 50 or fewer beds. 141 Figure 8 illustrates the distribution of public hospitals by size across states. This shows that, in 2013–14, the Northern Territory did not have any public hospitals with > 500 beds or ≤ 10 beds. Queensland had the highest number of hospitals with < 10 beds, and New South Wales the highest number of hospitals with 11–50 beds.

FIGURE 8.

Public acute and psychiatric hospitals, by hospital size, Australian states and territories, 2013–14. Based on Australian Institute of Health and Welfare material. 141

It is these small (district) hospitals in rural and remote areas that perhaps most closely resemble the traditional community hospital found in the UK:

[Community hospitals] are hospitals that perform less than 2000 separations [hospital discharges] a year and either have surgery or no surgery attached. So conceptually [this] includes most of our small rural towns, looking at populations between about 2000 and 10,000.

KI-A04

Community hospitals in this sense would typically be run by GPs, partly because ‘the specialists aren’t there’ (KI-A01); these types of hospitals are not common in metropolitan areas. Set-up and size will, however, vary by state. One key informant noted that the most common model in Queensland would be in the form of ‘GP-run hospitals where the GP you know provides anaesthetics and [does] basic surgery [. . .] There might be between 500 and a couple of a thousand people in a town or a district’ (KI-A02), while in New South Wales they would be somewhat bigger:

Probably around maybe 40 to 100 beds which are generally in smaller rural towns. And they provide surgery on low risk patients and do medicine, treat gastro[intestinal conditions], [intravenous] antibiotics, that sort of thing. A lot of them do chemotherapy as an example, a lot of them do renal dialysis; so they provide . . . they have what we call close observation units, so they don’t really provide intensive care.

KI-A06

One other key informant highlighted the specific situation in the state of Victoria, where public hospitals would have a comparatively high number of residential care beds, between 40 and 50, but only a small number (up to 10) of acute care beds, which are ‘mainly filled with people at the very low acuity categories and some of them might have as few as two (beds)’ (KI-A03).

One commonality observed by all key informants interviewed for this study is that the focus of community hospitals in the Australian context tends to be geographical rather than on specific population groups (KI-A01). However, boundaries may be blurred where the local community constitutes a specific group, such as indigenous populations. In addition, as rural communities are ageing, service delivery will have to accommodate the changing needs (KI-A06), as perhaps illustrated by the example of Victoria.

Types of community hospitals and services provided

Reflecting the purpose and settings of community hospitals in the Australian context, the range of services provided can vary across states and across regions within states. For example, in some settings, community hospitals would provide emergency stabilisation, management of simple conditions, hospitalisation for uncomplicated general medical patients, or subacute care or step-down care (KI-A04). Elsewhere, services would include obstetric care, emergency care and minor surgical procedures, which would be delivered by GPs (KI-A01):

GPs that are trained [. . .] in both anaesthetics and basic surgery so they might have a graduate diploma in anaesthetics so they are not trained anaesthetists but they do some general anaesthetic training. They also can do some surgical training, so they can do basic surgical procedures. So they can do you know severe infection, appendectomy, that sort of thing. So procedural GPs can do those, really it is around emergency surgical procedures.

KI-A02

Key informants commented that this model of ‘procedural GPs’ used to be quite common (KI-A02, KI-A03) but ‘waned over a period of time, as there were closures of small country hospitals across the country because of cost pressures, but now it has come back’ (KI-A02) as part of the rural generalist pathway (see also below).

Access to specialist expertise varies across settings and states, reflecting, in part, the governance models in place, as well as the level of collaboration and co-operation across the health system in individual states. For example, in Queensland, small public hospitals are managed by one of the 17 Hospital and Health Services (previously Health Service Districts) and the community hospital would be embedded within wider service provision (Box 3):

[T]here’d be a director of nursing at the local site or there may not even been a director of nursing, depending on what the industrial agreements say, but they’d be totally integrated in a management sense and clinically they would have built up referral relationships depending on where the hospitals are and very often there might also be telehealth links between the main hospital and the local hospital.

KI-A01

This would be somewhat different in the state of Victoria where hospital governance has been devolved so that most hospitals, including the smaller ones, will be their own organisation (KI-A03). However, hospitals tend to collaborate closely, with the regional hospital providing support for services that are not available in the smaller hospital:

[In] a local hospital, you’re not going to have an obstetrician or you’re not going to have even a permanent medical director, but you might essentially borrow that staff member from the larger hospital so that you can still have those services provided in your hospital without actually having to employ the person and that works quite well in some cases. It means [. . .] that there’s maybe somebody coming in once a month or whatever it might be who is overseen by somebody from a larger service or who is based in a larger service, so that it takes some of the clinical governance burden away from that smaller service and they don’t need to have a clinical governance framework for cancer care for example.

KI-A03

Key informants described similar examples in other states, such as the Northern Territory, where the local GPs would have a network of specialists whom they could telephone for advice or for evacuating patients if required and where ‘the specialists fly in and out, and you know run clinics or have surgical lists, that sort of thing’ (KI-A02) (Box 4). Such a model was also described for diagnostic services in small hospitals in New South Wales, where basic radiology or pathology services may be available on site, with the local GP and nursing staff functioning as ‘remote operators’ while a visiting clinician would provide the specialist expertise (KI-A04).

Several key informants highlighted the key role of telehealth technology in supporting the provision of services in rural and remote settings (KI-A01, KI-A02, KI-A04, KI-A06), although one expert interviewed for this study highlighted that the development of telehealth had not been progressed as far as it could have [‘We also have very poor internet infrastructure in Australia so that does not help’ (KI-A03)].

Benefits and disadvantages of community hospitals

There is some empirical evidence about the impacts of community hospitals on patient and service outcomes in the Australian context, but this tends to draw on small case studies analysing the feasibility of providing a given service and typically involving a before–after assessment of impact only. For example, two linked studies described the provision of emergency services in a small rural hospital in Victoria, which provided a 17-bed ward for medical, surgical and palliative inpatient care, radiography, ultrasound and pathology services and an on-site GP clinic, as well as three emergency department beds. 40,77 This found that the small size and lack of dedicated emergency department staff restricted the ability of the hospital to match the performance of major metropolitan emergency departments. This was seen to be particularly challenging in view of an ageing population and large proportion of after-hours presentations.

A small number of studies examined the use and impact of telehealth services such as teleophthalmology,44 rural videoconference fracture clinics50 or telepharmacy. 86 These found evidence of varying quality and robustness on the potential for telehealth technology to enhance patient and service outcomes, such as earlier diagnosis and treatment, a shorter hospital stay (where required) and avoided travel for an eye care service using internet consultation provided by a small regional hospital in Western Australia. 44 A reduction in unnecessary patient transfers to the tertiary referral centre was also reported for videoconference fracture clinics provided by a small rural hospital in Queensland. 50 The latter study also showed that, although the cost of running the videoconferencing clinic was initially higher than the costs saved because of avoided patient transfers, there was an overall saving for the health service in the second year of operating the clinic. A third study assessed the feasibility of implementing pharmaceutical review using videoconferencing between remote pharmacists and patients in two rural hospitals in Queensland, finding the service to lead to improved documentation of medications and reduced medication risk. 86

Aside from these studies, much of what is known about the benefits of small rural hospitals in Australia tends to be anecdotal:

Based on just my past experience and talking to people [. . .] So you know we do have an aging population, in many of these towns they have lost populations. There is no point in maintaining larger hospital facilities that are no longer needed whilst the population is aging and you know needing care. So I think that they are successful because they have actually been responsive to changes in the population.

KI-A02

One key informant highlighted the challenges of evaluating community-based programmes and service delivery models, because of their small size, ‘but there is a great deal of belief so you end up with this knowledge gap which is a challenge when we try and protect the funding of smaller services’ (KI-A03). What is known is that populations in rural and remote settings tend to have a higher utilisation of services, with evidence of higher rates of potentially avoidable deaths,225 although the degree to which this is attributable to the quality of service delivery locally remains poorly understood (KI-A04).

Challenges faced by community hospitals

Key informants interviewed for this study pointed to a number of challenges faced by community hospitals, and these related mostly to workforce and funding issues. There were differing views about the nature of workforce challenges, although there was agreement that rural and remote communities faced particular difficulties:

So especially if there are only a couple of GPs they can get quite isolated and the life can be quite difficult [on] 24/7 call, 7 days a week, 365 days a year and very often in these rural remote communities they might be staffed by international medical graduates who are, you know, have to work in a rural and remote area to get [the] full licence [. . .] So workforce issues are pretty dramatic in remote Australia.

KI-A01

There have been efforts to try to address these challenges, through, for example, the establishment of rural clinical and regional medical schools,226 with emerging evidence pointing to a positive impact in terms of attracting and retaining graduates to work in rural areas. 227 One key informant commented that this gave cause for optimism to ‘get on top of the workforce issue’ (KI-A02), while another key informant noted that, although there had been a ‘massive expansion in the number of medical students’, this was unlikely to address the challenge of attracting graduates to work in rural and remote settings ‘because most of the medical students are living in the cities and they’re not going to go and work in the country’ (KI-A01). This is in part because GPs can set up practice anywhere in the country, and most settle in inner cities and in the more attractive rural communities, and ‘there is nothing you can do to reallocate those GPs, you have to attract them to small rural communities’ (KI-A03).

The overall role of the GP varies across states and regions within states. One key informant noted that it was financially attractive for GPs to provide services in rural hospitals because general practices generate income through Medicare plus additional charging where they operate private practices (which is, however, not typically the case in very remote communities) (KI-A02). The state government, which funds hospitals, would then pay an additional stipend to the GP to run the hospital, as for example in Queensland:

[It] is a substantial amount that makes it worth their while to run the hospital. For a lot of them where general practice might not be as viable as it could be, that makes the big difference between you know make or break in maintaining their practice in that country town. So what they get from the state government is actually quite important to sustain those services.

KI-A02

In other states, small rural hospitals would not have a GP as clinical lead but it would be attended by a visiting medical officer. A GP cannot be formal lead because they are a private provider and, as such, the GP cannot be employed by the hospital (KI-A06). Instead, there would be a facility manager or a health service manager who would then contract a GP as visiting medical officer or some other form of agreement that would involve the GP providing clinics at the hospital on a fee-for-service basis. This situation can, however, become challenging in places where services are organised in a different way, such as in Victoria where many smaller hospitals operate as single organisations. One key informant noted that these hospitals will find it increasingly difficult to pay for after-hours services delivered by GPs, because ‘[GPs are] starting to increase the amount that they’re asking for beyond what the small hospitals are funded’ (KI-A03).

Overall, the funding situation is complex because of multiple funding sources, which is particularly challenging for small health services because they have to report back (on different performance indicators) to multiple funders (KI-A02):

[B]ecause we have a federal level of government and a state level of government, it is a very confusing policy space from the perspective of those small hospitals [. . .] There are things that are caused by our federal state system, like the fact that a hospital cannot employ a GP or finds it very difficult to contact with a GP because of the fact that they are federal rather than state, which sort of blurs some of what they do.

KI-A03

Future changes in community hospital provision

Small hospitals in rural and remote areas in Australia are facing several challenges, such as maintaining services and attracting staff, along with economic viability within constrained resources. This has led to an increase in closures of a number of small hospitals over the past decade (KI-A04):142

Rationally, they can’t operate [. . .] and the occupancy rates is so low, it’s very hard to argue that they’re operating efficiently and it’s increasingly evident that it’s hard to operate things safely, for anything with fairly low security levels.

KI-A05

However, at the same time, such facilities are seen to be indispensable in remote communities ‘where a patient can be stabilised and evacuated because the distances are significant’ (KI-A01), although this service might not necessarily be provided by a hospital as such. One key informant interviewed for this study noted that the general policy direction was to (re-)encourage the development of rural generalists, that is, GPs who have additional skills, such as in anaesthesia or surgery or obstetrics (KI-A01). Rural generalists (or rural generalist medicine) have been defined as practitioners who have the skills to provide unsupervised community or primary care, inpatient and emergency care in a hospital or related setting, extended specialised services in at least one approved medical discipline required to sustain comprehensive health-care services in regional, rural and remote communities, services across the continuum of care in a range of settings and service delivery models including outreach where required, such as community paediatrics or palliative care, and who are able to apply a population-health approach with relevance to the community in which they practice. 228 Training is provided, for example, through the Australian College of Rural and Remote Medicine. 228

There has also been a move, in most states, to transform small rural hospitals into alternative settings, such as residential aged care facilities, community medical centres or multipurpose health centres (KI-A05, KI-A06). This is illustrated by the example of Victoria, where public hospitals have a relatively high number of residential care beds, in effect representing residential aged care facilities, with community health visiting specialists (KI-A05). New South Wales has invested in the redevelopment of small rural hospitals into MPSs in response to an ageing population and the need to ensure appropriate care (KI-A06). This was part of the MPS Program, which was established in 1993. 229 A joint initiative of the federal and state governments, the services are designed to specifically address the needs of small rural and remote communities that could not viably support stand-alone hospitals or aged care homes. 143 Usually administered by the state or territory health department, MPSs bring together a range of health and residential aged care services on one site, with co-location of GP and ambulance services through pooling the funding across programme boundaries, so overcoming some of the challenges posed by the otherwise fragmenting system of health services and aged care funding. 230 They are available in all states except for the Australian Capital Territory:

It is a good model actually [. . .] It is a way of transforming small acute care facilities to meet acute, aged care residential care facilities. So that is quite common now and is a very successful model with the aging of the rural population.

KI-A02

Some will have inpatient beds or are co-located with a residential care facility. MPSs typically provide up to 10 beds, but they can be flexible and, depending on demand, beds can be redefined as acute beds or as respite beds to meet the needs of the local community (KI-A04). Available evidence has shown that the MPS model is responsive to community needs and able to tailor services to individual community needs. 143 The model allows sharing of staff among services, which may lead to the achievement of economies of scale through the amalgamation of acute and aged care services, which will be particularly important to enable small communities to sustain services.

These efforts form part of ongoing reforms across Australia to strengthen co-ordination between primary and hospital care as well as aged care to better meet the challenges of the rising burden of chronic disease and Australia’s ageing population. There are ongoing discussions of options for reforming the roles and responsibilities of the federal and the state governments in regulating and funding health services in Australia. These developments will probably impact on the future evolution of community hospitals in the country.

Defining the community hospital

In Finland, the term community hospital as used in the context of the NHS in the UK does not exist. However, there are service delivery models in place that resemble some of the key features of the traditional community hospital as conceptualised in the UK. The previous sections have shown that municipal health centres, which were established following the 1972 primary care reform in Finland, partly evolved from small local hospitals and that they offer a wide range of services, including GP-led inpatient units. 117 These are frequently referred to as health centre hospitals [‘We do not talk about community hospitals we talk of health centre hospitals’ (KI-F04)] or health centre inpatient wards and they provide short- and long-term (≥ 90 days) inpatient care. 252,253

There is no clear definition of the ‘health centre hospital’. The 1972 Primary Health Care Act describes the purpose of hospital beds provided at health centres, specifying that these beds:

Shall primarily be assigned to the patients that it is most expedient to care for [. . .] in view of the nature of the illness and the need for examinations, care and medical rehabilitation, or the degree of convalescence. A person in need of urgent institutional care shall always be assigned to a health centre hospital bed or, if the necessary examinations or care cannot be provided there, must be instructed to use, or consigned to, an appropriate medical care institution. 129

As noted, health centre hospitals partly evolved from the conversion of small local or district hospitals into municipal health centres in the context of the 1972 primary care reform. Subsequent to the reform, newly formed municipal health centres that were not linked to a former local or district hospital also established inpatient wards, for the reason:

That not all patients needing inpatient care are in need of specialised care led by specialists. Typically, frail, elderly, terminal care patients and so on, could quite as well be placed in smaller units closer to their home community, so that’s the explicit reason for that.

KI-F04

Care provided in health centre hospitals is considered inpatient care, as set out in the 2010 Health Care Act, while explicitly falling under primary care services. 182 It is for this reason that national data formally distinguish inpatient primary health care from inpatient specialised care. 238 Furthermore, a health centre may also comprise a specialist-led hospital. 257 This tends to be the case in larger cities that have a city hospital (KI-F04, KI-F05), which is administratively overseen by the municipal health centre ‘but the actual care given in those units is typical specialised care’ (KI-F04). This complicates interpreting national data on the utilisation of health centre inpatient care services at present.

For example, in 2014, just under 143,000 patients received health centre inpatient care, accounting for some 4.4 million care days. 257 Although the annual number of patients has remained at around 145,000 since the early 2000s, the number of care days has almost halved since then, as reflected in the average length of stay, which fell from just over 30 days in 2003 to 22.3 days in 2014. About two-thirds of patients were aged 75 years and older, accounting for about 75% of all health centre inpatient care days; those aged 85 years and over accounted for about one-third of patients and for about 40% of inpatient care days. The number of patients and of inpatient days varied by region, as does average length of stay, ranging from 11.3 days to 54 days.

The most common primary diagnosis for inpatients in health centres were circulatory diseases, with an average length of stay of 29 days, followed by injury and poisoning and other selected external causes (23 days), respiratory diseases (14.5 days) and mental and behavioural disorders (45 days).

As noted above, health centre hospitals offer short- and long-term care beds, although the latter form only a small proportion of the total number of patients in health centre hospitals. For example, in 2014, the number of patients aged 75 years and older residing in health centre hospitals as long-term inpatients was 2920 (3% of all patients in health centre hospitals), or 0.6% of the population in this age group, with proportions ranging from 0.2% in the Satakunta region in the south-west of Finland to 1.9% in North Karelia. 257 The proportion of older people receiving institutional care in health centres has fallen substantially since the early 2000s (Figure 12), in particular from 2009 onwards, reflecting the increasing trend away from institutionalised care towards the home-based care and sheltered accommodation alternatives mentioned earlier. 253 It is worth noting, however, that the proportion of older people receiving regular home care has changed only slightly during the 2000s, fluctuating between 11% and 12%,257 because the regular use of home care has tended to focus on those in greatest need. 253

FIGURE 12.

Long-term inpatients in health centres (as a proportion of the population in the relevant age group), 2000–14. Based on National Institute for Health and Welfare. 257

The total expenditure on inpatient primary care has declined from 6.8% of total health expenditure in 2000 to 5.8% in 2013, with much of the decline attributable to a large reduction in expenditure on long-term care, by 55%, while expenditure on short-term inpatient care rose from 3.2% to 3.8%. 238

Types of community hospitals and services provided

Health centre hospitals provide subacute and post-acute or intermediate care following discharge from an acute care hospital (KI-F01, KI-F02, KI-F05). Their role is to look after people who are not sufficiently unwell to be in an acute care hospital but who are clinically too unstable to be in their own home or the nursing home. The subacute function includes an entire spectrum of cases, from ‘simple’ infections requiring the administering of intravenous antibiotics to terminally ill people requiring palliative care (KI-F02). Health centre hospitals previously also provided 24/7 emergency services but this function has been transferred to general acute care hospitals during recent years (KI-F05).

Health centre hospitals are seen to be principally very similar in terms of the type of patients they take care of; the main difference is that of local context in terms of size of the community and the availability of other services provided (KI-F01, KI-F02):

Like the line between elderly persons’ homes and these inpatient health centres, so in some municipality’s maybe a more chronically ill patient is taken care of in the elderly persons’ homes and in some municipalities the same person would be taken care of at the [health centre] patient ward – the line can vary between municipalities, but not I think very much.

KI-F01

One key informant interviewed for this study highlighted the impact of the 1993 state subsidy for public services reform. As described above, it made local municipalities responsible for all secondary and tertiary care costs and this led to a change in the composition of patients cared for in health centre hospitals, and ‘they lost much of their professional attraction and gradually became long-term care facilities for the elderly, especially those with memory disorders’ (KI-F05).

Health centre hospital staffing would typically include a GP, a head nurse, physiotherapists, nurses, health-care assistants and, depending on the size of the municipality, social workers (KI-F02). In larger municipalities or cities, health centres would also have medical specialists such as in internal medicine and geriatrics available, and in places such as Helsinki neurology as well. The work of the GP can be exclusively on the ward, depending on the organisation and size of the health centre. For example, centres might divide work differently, having for example three GPs looking after patients on the ward while also taking care of regular health centre patients (KI-F02).

The lead physician of the health centre determines whether or not a patient requires admission to the health centre hospital as stipulated in the 1972 Primary Health Care Act; the lead physician is also in charge of determining any transfers of patients to other medical care institutions. 129 Admission can be from acute care, nursing homes or people’s own homes.

Given the origins of health centres as described above, the range of services provided would typically include the inpatient ward, an outpatient clinic, physiotherapy and a small radiography unit and very basic laboratory services (KI-F01) (Box 5). Some municipal health centres would also employ specialists such as a surgeon or a specialist in internal medicine. One key informant interviewed for this study pointed to the example of a municipal health centre in the very northern part of Finland, where the inpatient ward would be doing ‘a bit more than normal’ (KI-F01), because of its remote location.

At present, it is difficult to determine the number of beds in municipal health centres with certainty, and there appears to be no central guidance on the planning of beds to be hosted by health centre hospitals. Typically, there would be between 25 and 40 in one ward, but some health centres have two wards (KI-F02). The decision about size of the health centre and number of beds in inpatient wards is usually made by the municipal council (KI-F01, KI-F02).

Benefits and disadvantages of community hospitals

There is some empirical evidence on the impacts of community hospitals on processes and outcomes of care in Finland. A small number of studies examined the outcomes of people with COPD treated in primary care hospitals compared with secondary care hospitals. Lampela et al. 259 analysed hospitalisations for acute exacerbations of COPD among patients aged ≥ 45 years during the period 1995–2001 by specialty (general practice, pulmonary medicine, internal medicine). They found a general increase in the number of admissions during the observation period, of 11%, with the largest rise seen among those supervised by a GP (36.8%), suggesting a shift in the responsibility for COPD to primary care. This was attributed, in part, to a national programme for COPD and the related training provided for GPs, but also to financial motivations. Subsequently, the same authors specifically examined outcomes of COPD treatment by hospital site for the period 1995/6–2004. They found a reduced risk of death among those managed by a GP relative to those managed by a pulmonary specialist, at 0.83 (95% CI 0.75 to 0.91). 148 Conversely, a related study showed that COPD patients discharged from primary care hospitals had a greater risk of readmission within 7 days of discharge, compared with those discharged from secondary care. 149 The readmission risk was highest among those aged < 65 years, but not among those aged ≥ 85 years. The authors discussed a range of possible explanations for this observation, including differences in treatment procedures and arrangement of subsequent care, along with differences in disease severity (which they were unable to assess) and admission threshold, which may differ between the different levels of care.

Four key informants interviewed for this study generally believed that health centre hospitals could provide a useful ‘buffer between specialised healthcare and then nursing homes, residential homes or [sheltered] accommodation’ (KI-F01). The main benefit was seen to be the location of the health centre hospital close to people’s homes, which was considered to be an important feature given the geographic dispersion of the population in Finland, in particular for older people who can stay close to their relatives (KI-F02, KI-F05). One key informant commented that care provided in health centre hospitals can be seen to be more holistic or person-centred because the patients and their circumstances are known to health centre staff, as are the local services that are available to patients, such as home care services or sheltered accommodation. It can also be seen to be efficient in that they allow more specialised care provided by the hospital district ‘to concentrate on the real work, on the specialised care’ (KI-F02).

Challenges faced by community hospitals

Key informants also recognised the challenges posed by the current system:

We have also an amount of people with, I would say in the wrong place, because we don’t have always enough sheltered housing. So when the people need [. . .] during the whole day or 24 hours a day some kind of help and they have to wait for that kind of housing place, we have these also in there. That is the problem in the process.

KI-F02

It was noted that where health centre hospitals worked well, they were very successful in meeting their ambition of keeping people out of acute care hospitals, through for example investing in intensive rehabilitation for people following discharge from acute care (KI-F04, KI-F05):

Those advanced municipalities. So they have lots of rehabilitation type of activities for the elderly, in this unit. They know that if the specialised care hospital makes the elderly stay in bed for a week or two, that game is lost. So the more advanced municipalities try to invest in this type of rehabilitation wards, in the health centre hospitals.

KI-F04

However, key informants also highlighted that in several cases this was not achieved and, instead, that health centre hospitals had tended to develop into geriatric long-term care units, a ‘very long term geriatric patient repository type of thing which was not the idea’ (KI-F04).

There were some concerns among those interviewed for this study with regard to attracting health-care workers, especially in rural or remote areas in eastern and northern Finland with younger people moving away (KI-F01). However, the challenge to attract health-care workers has to be seen in the general context of workforce challenges in Finland (KI-F05):

[W]e have problems with finding enough doctors to work in the primary health-care centres. And to work on the ward it’s more demanding than in [outpatients], so that it is not so easy but because if you go to work to the health centre you are [a] worker there then the employer can decide that you are working in the open or in the ward. But of course, we always try to find people who really can and want to do those works because there are a lot of jobs to choose [from] so that if they say they have to go there then the workers say goodbye.

KI-F02

A recent increase in medical graduates is expected to overcome some of these challenges (KI-F05). Larger health centres may find it easier to recruit because they provide opportunity for GPs and nurses to work with specialists in internal medicine or geriatrics, for example (KI-F02). This may be particularly attractive for younger doctors given the characteristics of the patient population, that is, older patients with complex health and care problems. One key informant reported on experiences of ways to attract workers, highlighting the need to offer flexible terms of employment and allowing workers to, for example, ‘continue their studies or do research, or work in private sector also [. . .] It has to be some kind of a combination that this kind of a people and specialists are willing to work in a public sector and city hospital’ (KI-F03).

Future changes in community hospital provision

Key informants interviewed for this study agreed that the future of health centre hospitals should be seen in the context of the wider challenges faced by the health system in Finland:

One of the questions which is topical at the moment, and which should be discussed in all the Finnish municipalities, is how to develop the ambulatory specialised care, or, as I would say, how to abolish the division between primary and secondary care.

KI-F04

There was a perception among those interviewed for this study that health centre hospitals should be complementary to the specialist services provided for by the hospital district (central and university hospitals) to avoid duplication of services, because the municipality has to cover the associated cost either way (KI-F01). However, there may be a risk of duplication in regions where there is a city hospital, which is administratively overseen by the municipal health centre, and the central hospital provided by the hospital district (KI-F01, KI-F04). There was consensus about the probable impact of the ongoing government reforms, in particular for those municipal health centres that are operating specialised health-care services, and the need to cut these services, although perhaps not affecting those health centre hospitals that really provide GP-led care (KI-F04).

Earlier sections of this report noted that there has been a substantial fall in the proportion of older people receiving institutional care in health centres, in particular from the mid-2000s. One survey of municipal health centres carried out in 2013 noted that about 40% of the centres that had responded to the survey (141 of 151 approached) had reported changes in inpatient care, such as the closure or transition of services to other settings. 157 That study did not provide any further detail about the nature of these reported changes but this is likely to reflect, at least in part, the ongoing trend across Finland towards moving people out of institutional care into alternative settings. The main driver is seen to have been financial but the change was also seen to be motivated by the ‘ideology of promoting home care as long as possible’ (KI-F05). One key informant interviewed for this study questioned the success of these policies and noted:

So one thing municipalities actually do is they’ll change the name of the nursing home to [sheltered] housing, everything stays the same, but it’s a positive step for the municipality and social institutions will pay the bills [. . .] there won’t be any change for the people living there.

KI-F01

This statement is somewhat in contrast to the perception of one other key informant from one municipality that has begun moving patients from the health centre hospital to sheltered accommodation who reported anecdotal evidence that the quality of life had improved for those concerned:

Not in every case, but it might be with the majority of the patients [. . .] They have a lot of open possibilities for different kind of activities. They have their own rooms and so on, so I guess we are sure that we have a better quality in [sheltered accommodation] than how the hospital works’.

KI-F03

It is not possible, on the basis of these two contrasting views, to arrive at an overall conclusion about the best way forward, although there was consensus among all key informants interviewed for this study that the organisation of specialist services in the present system would need to change, in particular those provided in smaller hospitals, because of their perceived lower quality (KI-F01). There was also recognition, however, that this is going to take ‘a long time because it’s the small municipalities that have these regional hospitals and they don’t like the idea that it’s going to be taken away from them’ (KI-F01). Furthermore, where health centre hospitals are closed altogether this would require support for workers, especially nurses who would be expected to work in a different environment (home care services, city or regional hospital), and providing this type of support should form a key part of the ongoing reform efforts (KI-F-03).

There was consensus among key informants that the ongoing reforms could provide considerable opportunities to rethink the processes by which health care is being delivered. Key informants (KI-F02, KI-F04) saw the reform to provide an opportunity to enhance seamless care between primary care and specialist care that will then eventually also help to reduce the number of beds in health centre hospitals:

And I’m sure that after 10 years we still have those wards in smaller, not in so many municipalities as we have now. But we are going to have them but I’m sure we are also having much [. . .] better processes [. . .] I hope there are no more so strict borders between primary health care and specialised health care and we are concentrating more on the process and during the process patient’s needs different kind of professionals.

KI-F02

They also noted that these sorts of changes needed to be developed in a measured way, rather than forcing the reforms through too fast in order to achieve sustainable transformation.

Defining the community hospital

In Italy, the development of community hospitals (Ospedale di Comunità) has been described as ‘a very fragmented experience’ (KI-I02). They were set up from the mid-1990s onwards, with the Emilia-Romagna region being the first to introduce such structures in the provinces of Premilcuore (1995) and Modigliana (1996). 138 Adapted from models in place in the UK, the concept of the community hospital in Italy spread steadily in a number of regions, including, as well as Emilia-Romagna, the regions of Tuscany and Marche, followed by Umbria, Puglia, Friuli Venezia Giulia, Piedmont and Lazio, with the majority having incorporated community hospitals into their regional health legislation. 137

A common feature of the early community hospitals in Italy was their ‘bottom up’ evolution in the absence of a national framework on how to address the rising need for integrated health and social care planning and provision. 138 Initially also referred to as ‘country hospital’, the Italian National Syndicate of Territorial Doctors described this new structure as ‘a territorial social health facility, developed in response to the health needs of those who do not require, in the view of the family doctor, the complexity of secondary care but at the same time whose health and social care needs cannot, for various reasons, be adequately addressed in his/her home environment’. 276 Subsequently, the Italian Federation of General Practitioners offered, in 2003, a refined definition, which describes the community hospital as a regional health structure within the network of district services, which:

• provides for the direct management of beds by GPs of that catchment area, aimed at service users, mainly dependent older people, who do not require hospitalisation but who cannot be cared for in their own homes

• is managed by GPs and allows for person-centred care and medium- to low-level clinical interventions appropriate for the management of chronic degenerative diseases and people with a range of social care needs

• is placed at the interface (‘buffer’ or ‘bridge’) between the hospital and the community through a network of home care services and nursing homes.

A review of the early experiences of community hospitals in Italy presented in 2003 noted that their nature and scope varied considerably, reflecting to a great extent the local context in which they were implemented. 138 It further highlighted that the flexibility of the community hospital model was to be seen to be one of its major advantages and was the result of the absence of a rigid legislative framework. This permitted multiple experimentations but has posed challenges to the systematic comparison of different models. Despite this heterogeneity, the review identified three basic models:

1. The ‘rural model’ (modello rurale): defined by the hourly availability of a defined number of GPs who individually agree to be present for a set number of hours/time slots in the facility to provide care where required. GPs are remunerated by a set daily fee against specific goals of care that they are committed to achieve. Bed numbers vary between 5 and 8 beds up to a maximum of 12–15 beds. As is typical for small and medium-sized towns, the model was most commonly found in Emilia-Romagna and Marche.

2. In the ‘metropolitan model’ (modello metropolitano), the GPs sign up with the facility on the basis of a programme that is agreed in advance (with the district director, etc.) and with fixed dates (weekly, biweekly, monthly) to monitor inpatients. Usually found in urban centres, where the number of doctors who can provide care is high, the model varies widely, with the number of beds ranging from 8 to 12 up to a few tens. It appeared to be the most common approach in Tuscany, Umbria and Friuli Venezia Giulia.

3. In the joint (misto) model, the clinical responsibility of GPs is shared with specialists or other staff. The model provides for the simultaneous presence of the GP (access or time slot) and medical staff in various forms of integration. It was most common in Marche and Umbria.

A further model that did emerge in some parts of Italy was described as the ‘hospital model’, which, in contrast to the community hospital that is normally managed by a GP, is managed by hospital doctors.

The first formal mention of the community hospital in national-level strategies came only in the national health plan 2006–8, which, in the context of a proposed reorganisation of primary care, highlighted the need for the development of intermediate care services and of GP-led community hospitals. 139 This notion of a community hospital as a structure within the Italian health-care system was only recently strengthened further, following the aforementioned Law 135/2012 on the reorganisation of hospital care, which envisaged that community hospitals to would play an important part in the delivery of more integrated services at the local level. This was further detailed in subsequent regulations (Box 7), and the 2014–16 health agreement described above requires the establishment of community hospitals for the delivery of subacute care, with the overarching aim of reducing the number of inappropriate admissions to acute care hospitals and facilitating the discharge of patients from acute care. 140

The number of community hospitals in Italy slowly increased over time, from 23 in 2003138 to 57 in 2009. 137 Distributed across eleven regions (Emilia-Romagna, Tuscany, Marche, Umbria, Puglia, Friuli Venezia Giulia, Piedmont, Lazio, Liguria, Veneto, Lombardy), some 200 municipalities were involved in the provision of community hospitals in 2009, covering a population of about 2 million. The total number of beds provided was about 700, which were overseen by an estimated 500 GPs and at least 1000 nurses and other health-care workers. It is difficult to arrive at more recent estimates for the number of community hospitals across Italy. This is in part because of the continued evolution of the role of the community hospital in the local context, and, perhaps more importantly and as we shall see below, because of the 2015 national legislation, which required regions to implement the stipulations set out in the national framework. 107

Types of community hospitals and services provided

Despite the recent formal definition of a community hospital in the national regulation, it remains difficult to map this conceptualisation to facilities ‘on the ground’ (KI-I01). One way of describing community hospitals in Italy is by identifying what they are not. Thus, the community hospital is generally not seen to be a long-term care institution for the chronically ill. This function is fulfilled by the RSAs, which have wards for seriously chronically ill patients. Community hospitals are not hospices either, as they are not designed to admit patients in the very terminal phase of their life. However, boundaries are not clear cut, with some community hospitals hosting or co-located with RSAs or providing a hospice function.

One key informant suggested differentiating two principal types or functions of community hospitals in the contemporary health system in Italy, namely subacute and post-acute functions (KI-I01):

1. Subacute: targeted at people aged ≥ 75 years with multiple conditions, who require (nursing) assistance that cannot be provided by their informal carers in their own homes; aimed at preventing admission to acute care hospital. They tend to be more common in rural areas.

2. Post-acute: intermediate care delivered in small hospitals or in separate units in larger hospitals; nurse-led, with support from health-care assistants; aimed at reducing length of stay in acute care hospitals. Post-acute care units or facilities tend to be located in urban areas.

A further form of community hospital are long-stay (lungodegenza)/rehabilitation beds, a combination of long-term care and geriatric rehabilitation with an average length of stay of about 60 days, which aims to stabilise the patient and support the family in the transition of the patient to a nursing home.

It is beyond the scope of this study to present an inventory of the wide range of community hospitals that have been established across Italy. Instead, we report on a small set of regions that have had experiences of community hospitals and for which publicly available documentation allowed for an assessment of the most recent developments. These are Emilia Romagna, Veneto, Marche and Friuli Venezia Giulia. Tuscany, which was also among the regions with experiences in developing community hospital-type structures, pursued, from 2013, a reorganisation of the regional health service, which foresaw, among other things, the integration of primary care with intermediate and hospital care of medium to low intensity, by means of establishing houses of health and the use of community hospitals, RSA and hospices as locations for intermediate care delivery. 277 However, from January 2016, the Tuscan health system has undergone a major reform, involving the merger of its 12 ASLs into three larger co-ordinating authorities (area vasta), which will further impact the organisation of regional health services. 278

Benefits and disadvantages of community hospitals

There is very little empirical evidence on the impacts of community hospitals on processes and outcomes of care in Italy. Much of what is known appears to be anecdotal and based on presumed benefits, with one key informant interviewed for this study citing the rising number of people aged ≥ 75 years who are surviving with chronic disease as indirect evidence of the impact of improvements in intermediate care (KI-I04).

There appeared to be agreement among interviewees about the implicit benefit of community hospitals, as they relate to their size, their closeness to the community and the active involvement of patients and their families and carers in the recovery process:

The other thing is [they provide opportunity] to train the patient, or the relatives, to use some facilities, where they have a rehabilitation programme, so use wheelchairs, or other things [. . .] So I think that this is quite a good place, it’s not so crowded, not too many beds, and also where the organisation is different from the hospital. Where the people can be referred and the relatives can understand what they need to do after.

KI-I03

Alongside these perceived benefits, study participants also noted a number of challenges faced by regions in developing community hospitals as part of a more integrated system locally. One of the main challenges described by key informants is for regions to have to strike a balance between implementing the national level norms on the number of beds and length of stay in acute care in particular [‘You cannot stay in acute, a real acute hospital more than seven days and then you are out. . . We don’t want to see you any more’ (KI-I-01)], against an increasing need for services because of the combined pressures arising from multiple chronic diseases and population ageing:

First of all a very big push for reducing the beds in the hospitals, as you know we have a regional standard [. . .] suddenly now we are down to three, three beds for 1000 persons. We were [. . .] at [bed] occupancy of 75% and now they require 90% of occupancy. And we had the pressure on the emergency rooms, so all the territorial structures are created first of all by looking at the problems of the hospitals [. . .] but if you don’t resolve the problem of the territorial assistance, you will never be able to keep the people out of the hospitals.

KI-I02

This key informant further noted that there was no ‘clear definition of what the continuity of care should be at a territorial level’, combined with a perceived lack of a ‘clear capacity of stewardship if you want from the national level’ (KI-I02), leading the regions to take different approaches, which has implications for the accessibility and performance of health care across the country, a challenge that was also mentioned to be a wider challenge for the Italian health system.

Against this background, the way the systems at local or regional level have evolved can be seen to be opportunistic rather than needs-based, driven by the need to reduce capacity in the acute care sector, which may impact on the delivery of appropriate care to those in need:

So you may have in one local health authority twice as much subacute as the other one, the neighbour local health authority, who has half of these and twice of those. Or three times nursing beds or four times this here, so very random. So, what we have found is that, the sum of all this kind of beds are moving towards an increasing number because of the reduction of [acute] hospital beds and because the development of nursing homes, paid by families. But the combination is quite emergent, with no real rationale [. . .] But having such big differences in the mix, in the infrastructure mix, what happens is that, independent of your physical or psychological condition, you can end in one of those settings just depending on the offer you find over there.

KI-I-01

This challenge was seen to be further compounded by staffing policies introduced by government in the wake of the 2008 financial crisis. In brief, following the crisis there was a push to cut capacity among health-care workers, with national expenditure on the public health workforce falling by about 5% between 2011 and 2014. 155 Many of the ‘cuts’ in the physician workforce have been the result of a freeze on the restaffing of vacant posts following retirement, with a complete freeze in regions subject to central government-imposed regional recovery plans (i.e. regions with financial deficits in health-care budgets, mostly those in the south of Italy), whereas stronger regions that were in fiscal balance were permitted to restaff 50% of retired staff (typically in northern Italy).

This has created challenges, as the need for services is increasing; however, there are restrictions on replacing staff who have left alongside a need ‘to continue with the job, because if you have beds and you need to ensure 24 hour [care] you need to have at least six nurses, six nurses in a day, because you need to [cover services 24/7]’ (KI-I03).

As well as these broader pressures, one key informant also highlighted more local challenges around collaboration and integration, both within and between the health and social care sectors. For example, districts are generally responsible for ensuring that community hospitals have access to specialist expertise, which is typically based in the acute care hospital. And although community hospitals might have a list of specialists to call upon for a second opinion, this might not be the same the other way round and, as a consequence

It is possible we lose some patients now because doctors and nurses are in acute care level [do] not know there is [a] country hospital near the hospital or in the same district. But I think hospital doctors and hospital nurses need to have another thing after acute care because acute care is so, so fast now.

KI-I04

The same key informant also highlighted the challenges of working across sectors, including the introduction of social workers in the intermediate care team, while recognising that time had been among the most important factors that had helped to facilitate this collaboration:

So they are used not to working with the health-care system. They add a further professional in the unit, evaluation unit for elderly, for disabled, for access to the social factor and they go also in hospitals to evaluate people . . . Now it is more developed than the start because it was really difficult the first years, but now no they have used to work with us and we are used to working with them.

KI-I04

Future changes in community hospital provision

Community hospitals in Italy emerged as one of the possible responses to national policies that sought to focus hospital services on acute care and to reduce length of stay. They are seen to provide a means by which to prevent the hospitalisation of a patient or the provision of a ‘protected place’ following discharge because the patient is not fully stabilised or does not have appropriate support in their own home.

The development of community hospitals can also be seen as a policy response to the decision to close hospitals. Traditionally, municipalities have strongly opposed hospital closure in their territories, even if these performed poorly. Community hospitals provide one option to retain a health-care facility in small to medium-sized municipalities. Indeed, as we have seen, several regions have ‘closed’ old local hospitals by means of restructuring and reorganising these into multifunction facilities: ‘So we save the small local hospital, we keep it over there. We try to name it hospital . . . its written in big [letters] . . . so they [patients] feel comfortable to go there, but I mean no expert would ever name it a hospital’ (KI-I-01). These buildings may host a nursing home, a community hospital, specialist clinics and GP group practices.

The future evolution of community hospital provision in Italy can be seen to continue to be driven to a great degree by the national regulation of July 2015, as highlighted in a number of recent regional reorganisation activities targeting the hospital sector in particular.

Defining the community hospital

In Norway, the concept of a community hospital is not well defined and, among those interviewed for this study, some struggled with the idea itself and the difficulty to translate it into Norwegian (KI-N01, KI-N05). Some interview participants compared community hospitals with the traditional cottage hospitals in rural areas, especially in the northern parts of Norway (KI-N05), while others equated them to what is referred to as teaching nursing homes (KI-N06).

One approach to defining a community hospital in the Norwegian context is through its function within the health system, as there are several types of organisational structures that serve a similar purpose. More specifically, community hospital-type structures are generally understood as those that provide subacute and post-acute care (also referred to as step-up and step-down care) with the overall aim of increasing hospital efficiency, through preventing unnecessary admission to an acute care hospital, while providing care as close as possible to the patient home. Following this line of reasoning, there are essentially four basic structures that can be distinguished within the Norwegian system that fulfil this function and that we will describe in turn:

1. cottage hospitals

2. small general hospitals

3. municipal acute care beds (KAD)

4. local medical centres.

Types of community hospitals and services provided

Benefits and disadvantages of community hospitals

There is some evidence from Norway examining the impact of community hospitals on service and patient outcomes. For example, Aaraas et al.,313 in an early study of admissions to what they referred to as GP hospitals in Finnmark county, found that admissions to community hospitals can lead to lower readmission rates and cost savings compared with general hospital admissions. More recently, evidence from Hallingdal hospital, described above, suggests that of patients initially admitted to the community hospital, only 17% were subsequently transferred to the general hospital for further treatment. 21 Although the study did not use a comparative design, the authors argued that the community hospital was able to resolve the vast majority of acute care admissions (> 80% during the observation period) and therefore demonstrated its role as a ‘buffer’, protecting the general hospital from unnecessary admissions while filtering out those patients in need of further diagnostic work or treatment. A subsequent study found that patients admitted to the community hospital reported statistically significant greater satisfaction with several care aspects than those admitted to the general hospital. 150 Specific factors that were highlighted by patients included the ‘quiet and homelike atmosphere’ of the hospital, its close ties to the local community and a perceived continuity in the patient–staff relationship.

Garåsen et al. 9 examined more specifically the benefits of intermediate care in the Norwegian setting, exploring the experience of the Trondheim teaching nursing home. Using a RCT design, they showed that among patients allocated to intermediate care provided by Trondheim teaching nursing home, a significantly smaller proportion (19.4%) was readmitted for the same condition than for admissions to the general hospital (35.7%). Furthermore, after 26 weeks, among those initially admitted to intermediate care, 25% did not require community care compared with 10% in the general hospital group. On average, treatment costs in the intermediate care group were, at almost half the cost, significantly lower than for the general hospital group, at NOK 39,650 and NOK 73,417, respectively. 62 However, there were no significant differences in the number of patients who had died at 12 months’ follow-up. 62

More recently, Swanson and Hagen147 examined early evidence of benefit that can be associated with municipal acute care beds. Using a retrospective before–after cohort study, they found a correlation between the introduction of these beds and a small but significant reduction in overall hospital admissions (–1.3%, p = 0.041 for the general population and –1.7%, p = 0.041 for those aged ≥ 80 years). In particular, acute admissions and acute internal medicine admissions fell significantly for all age categories (–3.6%, p < 0.01 and –3.5%, p < 0.01 for the general population; –3%, p < 0.01 and –3.1%, p < 0.01 for patients aged 70–79 years; and –2.9%, p < 0.01 and –2.8%, p < 0.05 for patients aged ≥ 80 years). Based on these findings, the authors concluded that municipal acute care beds are likely to have alleviated the burden on hospitals by reducing acute care admissions. 147

In addition to the empirical evidence, those interviewed for this study pointed to a number of additional benefits for patients, citing anecdotal evidence in particular in relation to older people, who make up a large proportion of community hospitals’ patients:

Patients report greater satisfaction [. . .] And other factors enhanced are this interdisciplinary approach, the holistic approach, proximity to local GPs and the local health and care system but also a close co-operation with specialist health services at the hospital.

KI-N05

Another interview participant also noted the benefits associated with reduced distances to travel for treatment and its importance for older people in particular (KI-N01).

Challenges faced by community hospitals

Despite the observed and perceived benefits of community hospitals, there were reports of increasing professional objections to acute admissions of older patients to community hospitals, with concerns over the medical challenge that those with multiple health problems and complex needs, in particular, represent. 314 Key informants commented on the difficulty over deciding whether or not a patient qualifies for a community hospital admission (KI-N02), a responsibility that sits with the GPs: ‘It’s a difficult decision for the referring GP to decide whether the patient should go one place or another place.’ (KI-N05).

Several study participants also highlighted issues around skills, training and staff recruitment (KI-N01, KI-N02, KI-N04), with one key informant emphasising the need for GPs to attend ‘extra specialist training’ (KI-N01) to be able to treat complex and acute cases, and another participant pointing to the need to provide additional training to ambulance staff, who will have to deal with an ever-growing number of emergencies locally (KI-N04). Workforce issues were recently acknowledged by the government, which proposed to rethink hospital staffing and medical training in the context of the recent health service changes. 315

Interview participants also expressed concerns about how the various service innovations that have been promoted following the 2012 Care Coordination reform sit alongside each other, in particular with reference to the newly introduced municipal acute care beds (KAD). One study participant noted that municipalities would require more guidance from the state to support them in implementing these new models of care (KI-N01). This concern was also highlighted in a recent review of the quality of care in Norway. 289 It noted that, in order for the new vision introduced by the reform to be implemented fully, it was necessary to ensure that basic structures to support changes are in place, including appropriate payment systems and meaningful engagement between key stakeholders.

Future changes in community hospital provision

Increasing financial pressure and growing health-care needs are contributing to shaping the future of community hospital provision in Norway. The role of community hospital is reported to be changing, from small cottage hospitals providing care to remote population to a (hopefully) cost-effective alternative to specialist health care that can manage the needs of complex older patients through integrating all parts of the health system (e.g. local health centres, nursing homes) (KI-N05). Therefore, community hospitals are bound to become an even more diverse and flexible concept, with structures that are co-financed and co-managed by a range of stakeholders who will have the autonomy to design local agreements and collaborations. 154 Community hospitals will have to find their place in an evolving system in which the centralisation of more specialised services (especially acute surgery) and the investment in intermediate care services in local hospitals will be concomitant. This shift towards the development of intermediate care facilities under the authority of the municipalities is likely to create some tensions: local providers of specialist care such as the cottage hospitals and small general hospitals will have to compete not only with each other but also with newer facilities such as the local medical centres for funding (KI-N03).

Defining the community hospital

Many community hospitals in Scotland pre-date the formation of the NHS. 13 As such, these hospitals developed independently, often established by local benefactors in response to the needs of the local community. Thus, some developed from early fever hospitals or war memorial hospitals,13 whereas others evolved from local authority facilities that provided care for older people or people with mental health problems (KI-S01). By 1980, a study estimated that approximately 70 hospitals could be classified as GP hospitals, estimated to contain 3% of all hospital beds in Scotland. 135 According to the Royal College of General Practitioners, by 1990, there were 73 community hospitals in Scotland. 91

Against this background, it is perhaps not surprising that defining a community hospital in the Scottish (and the wider UK) context has remained challenging. A commonly used definition describes the community hospital as:

A local hospital, unit or centre providing an appropriate range and format of accessible health-care facilities and resources. Medical care is normally led by GPs, in liaison with consultant, nursing and allied health professional colleagues as necessary and may also incorporate consultant long stay beds, primary care nurse-led and midwife services. 104

However, there is no official definition adopted by the Scottish Government, and while ISD Scotland lists 60 community hospitals,331 the Community Hospitals Improvement Network lists around 80 community hospitals (KI-S03) as well as a proposed definition (Box 11). Since early 2015, the Community Hospital Improvement Network has been working together with ISD Scotland and four NHS Boards to agree a revised definition of a Community Hospital (KI-S03, KI-S04).

Types of community hospitals and services provided

Services provided at community hospitals vary widely and can include not only inpatient facilities, but also outpatient, diagnostic, day care, primary care and outreach services for patients, provided by multidisciplinary teams. 1 In response to technological advances and demographic pressures, community hospitals are reported to be expanding the range of outpatient and inpatient services, diagnostics and procedures that they can undertake. 13

The 2012 Community Hospitals Strategy Refresh by the Scottish Government sets out three main functions of community hospitals in Scotland:13

1. supporting the rehabilitation and recovery of patients after a stay in an acute hospital

2. providing the specialist end of community services, whether this be co-ordinating care around high-risk patients, providing/co-locating services that patients have traditionally had to travel to bigger sites for and beds for clinical specialties (such as palliative care and mental health)

3. providing some diagnostic and outpatient services.

However, many community hospitals provide further services, for example, dementia care, diagnostic tests and procedures, minor injury services or general clinical services. In addition, the co-location of other services on community hospital sites has increased in recent years, including social care practitioners, minor injury units and the Scottish Ambulance Service. 13

The 2015 Community Hospital Improvement Network working definition of community hospitals (Box 12) suggests a number of services that are likely to be provided within a community hospital, and key informants interviewed for this study were largely in agreement about the types of services that might be offered. 130 There was some divergence of opinion, however, with some key informants suggesting that all community hospitals should have inpatient beds (KI-S03), but others suggesting that a less restrictive definition might be more appropriate as community hospitals continue to evolve (KI-S02).

Within the overall community hospital definition, a number of different models of community hospital were noted by key informants:

1. non-bypass community hospitals

2. extended role community hospitals

3. long-stay geriatric hospitals.

Non-bypass community hospitals are those that are typically more than 1 hour’s journey time from an acute hospital, and they provide a wider range of services. These include 24-hour a day, 7-day a week A&E services to stabilise patients in need of emergency care prior to transfer to the acute hospital (KI-S03). 130

Extended role community hospitals provide a larger range of services than the typical community hospital (e.g. more complex surgery), as they have typically developed from a previous district general hospital and maintain a larger site. The hospital often changed to a community hospital function following the development of a new acute hospital nearby (KI-S03).

Long-stay geriatric hospitals, which have primarily long-stay beds and may be led by a consultant rather than a GP, may also be considered under the community hospital definition, although there was no consensus among those interviewed for this study on this point (KI-S03, KI-S05).

Community hospitals in Scotland were seen to sit ‘in the middle of the patient journey between home and the district general hospital’ (KI-S03) and have been described as an ‘integrator’ as well as a location for the development of a ‘single point of access to integrated services’ (KI-S02). It has been suggested that discharge of patients for rehabilitation to a community hospital can provide better integration with community and social care services. 13 In addition to their direct provision of inpatient services, community hospitals often host a wide range of different service providers on the hospital site, which may include GP practices, dentistry, community nursing, allied health professionals, social work, social care services and local council staff (KI-S03).

So you’ve got your district nursing base, usually at the community hospital, occupational therapy, physiotherapy base at the community hospital. Some have got dentistry. Some do minor surgery. You’ve got social work based there, children’s services, and in terms of somewhere to put staff within their local area, it’s a very good facility. And they also form out-of-hours hubs as well in a lot of cases.

KI-S05

Of the key informants interviewed for this study, several commented on the ongoing development of the integrative role of community hospitals, articulating an opportunity for community hospitals to develop into ‘local resource centres’ working within the broader health and social care environment as a hub for further non-inpatient activities (KI-S01).

Community hospitals are presently managed and owned by the local NHS Board. The NHS Board is responsible for funding recurrent and capital costs of the community hospital, balancing priorities with funding for acute care and other community care. Key informants suggested that changes in the organisation of health service delivery as a result of the 2014 Public Bodies (Joint Working) (Scotland) Act160 will mean that, from 2016, strategic plans will be developed by Integrated Joint Boards and wider decisions about funding commitments may be taken at this level (KI-S02, KI-S01).

Key informants further noted that local communities would feel a strong sense of ownership for their local community hospital. This sense of ownership was seen to be very powerful and particularly manifested in situations in which there are proposals to remove services or close a community hospital: ‘If you say to an area, “We’re going to get rid of your community hospital,” well, it’s all hell breaks loose’ (KI-S03).

Community hospitals are staffed predominantly by nurses and allied health professionals who are employed directly by the Health Board and who deliver the majority of patient care. 13 Medical leadership is generally provided by one or several GPs, and GP contracts for community hospital work are being negotiated locally between the GPs and the relevant Health Board (KI-S03). Patient care at the community hospital is typically provided by primary care multidisciplinary teams, with support from secondary care consultants as necessary. 13 Outpatient clinics may also be provided by visiting specialists from the acute hospital (KI-S04).

Benefits and disadvantages of community hospitals

There is little empirical evidence for the effectiveness and efficiency of community hospitals in Scotland. One study conducted in the Highlands found that general practices with access to GP beds used substantially fewer district general hospital bed-days per 1000 population than general practices without access to GP beds. 332 One key informant interviewed for this study reported on an unpublished evaluation of two GP practices, one with and one without a community hospital, which suggested that the community hospital practice was able to decrease overall costs of care through a reduction in outpatient referrals, A&E admissions and emergency admissions, with cost per bed at the community hospital reported to be lower than that at the acute hospital (KI-S05). Despite the scarcity of published empirical evidence, key informants cited a wide range of perceived benefits for patients cared for in community hospitals as opposed to acute hospitals, referring to a ‘calmer’, ‘more homely’ and ‘enabling’ environment, which may allow for better recovery of confidence and skills after an episode of acute illness (KI-S03, KI-S02, KI-S04). They also cited a shorter travelling distance for urgent care or outpatient appointments as a benefit (KI-S02), particularly for those frail older people who may find such travel exhausting (KI-S03), alongside increased ability for friends and relatives to visit (who may also be older) (KI-S03, KI4), facilitated discharge planning owing to ease of liaison with community services (KI-S02, KI-S05, KI-S04), reduced risk of hospital-acquired infection and of delirium and falls (KI-S02), reduced overinvestigation or overtreatment of patients (KI-S05) and better-quality food (KI-S05). One key informant further observed:

[The community hospital is] seen as part of a local support system for people, particularly but not uniquely perhaps older people or those who may require additional support. So I think that there’s a kind of intangible value there. It’s actually quite hard to demonstrate the . . . the value in, kind of, cash terms but in terms of emotional support as well as the physical support which is given, these local facilities have very significant value.

KI-S01

Key informants were also asked about perceived disadvantages of community hospitals, and the only observation cited was that some severely ill patients would require transfer to the acute hospital, although this was seen as less of a disadvantage of the community hospital as such compared with spending the whole hospital stay in the acute hospital.

Challenges faced by community hospitals

One main challenge noted for the delivery of care in community hospitals was seen in the difficulty of staffing community hospitals (KI-S03), along with workforce capability and skills. Despite a perceived high retention of the community hospital workforce, shortages in GPs and nurses, and the priority given to staffing of acute hospitals, have meant that community hospitals can become short-staffed (KI-S05, KI-S04). In addition, it was reported that it takes some time for new staff, who have received standard nursing or GP training, to develop the additional skills required for working in a community hospital. As one key informant stated:

It takes a couple of years, to be honest, to get people from having gone through a standard GP training programme to being comfortable about working here.

KI-S05

There have been some developments recently in the provision of further training for GPs. As an example, an additional year of community hospital training as an optional addition following the completion of the standard GP training is soon to be piloted (KI-S01, KI-S03, KI-S04). Other schemes also exist, for example Rural Fellowships in General Practice. 153 Other than these special schemes, it was reported that most of those going to work in community hospitals will learn on the job, through intensive mentoring by senior staff (KI-S05).

This need for additional workforce development may be part of a pattern that is changing at a national level. A 2013 UK-wide report identified a need for change in the medical workforce in response to changing patient needs in order to better respond to the rising burden of chronic disease, in particular multiple comorbidities. 333 It suggested that more generalist doctors will be needed who will be able to work in community-based multiprofessional health and social care teams and proposed changes in training pathways to facilitate this.

A further challenge that was reported to have emerged over the past few years was a perceived rise in the use of the community hospital as a step-down facility from the acute hospital (KI-S04) owing to pressures to free bed use in acute hospitals. This was seen to have reduced the number of beds available for GP admission (KI-S05). Furthermore, key informants noted that there had been delays in discharge from the community hospital to community care because of a lack of care home placements or the availability of home carers, which in turn would lead to longer length of stay at the community hospital than was seen to be appropriate (KI-S05, KI-S04).

Other challenges reported for community hospitals included the age of the infrastructure of some community hospitals (KI-S03), swift access to diagnostics (KI-S02), as well as constraints on health-care spending (KI-S02), which were seen to pose a challenge for the entire health-care system.

Future changes in community hospital provision

In 2006, a report from the Scottish Executive set out a strategy to refocus community hospitals, to contribute towards a shift in the balance of care from an acute setting to a primary and community setting and to an increasing emphasis on patient-centred care. 31 The report recommended that ‘NHS Boards should consider using new community hospitals as local resource centres in which to provide people with more holistic and integrated services quicker and closer to home’. 31 More recently, the 2012 Community Hospitals Strategy Refresh set out a number of suggestions for how community hospitals could improve patient outcomes. It suggested that community hospitals needed to be better integrated into the wider care provision, including acute hospitals, care homes and social care services. 13 Community hospitals were proposed to play a key part in the health and social care integration reforms, acting as a base for community services, or as a community hub for health and social care services. As we have seen, community hospitals are already providing an integrative role, and key informants interviewed for this study suggested that this role would develop in the future, with the potential of the community hospital to function as an integrated health and care facility for the community, hosting a range of statutory and non-statutory agencies (KI-S01). However, it was highlighted that the community hospital model needed to remain flexible to be able to respond to the diversity of local communities and their needs (KI-S01).

Key informants suggested that the ongoing aim of provision of care as close to home as possible, as set out in the 2020 vision for Scotland136 (KI-S02, KI-S03), had strengthened the role of the community hospital in recent years, given their ability to provide hospital care closer to people’s homes.

Key informants reported changes in the function of community hospitals in recent decades, particularly a move towards the use of community hospital beds for shorter-term acute stays, moving away from housing long-stay patients (KI-S03, KI-S01). One exception to this is dementia care and palliative care, where community hospitals may continue to provide care for a longer period (KI-S01). It was also reported that community hospitals now receive patients who are more acutely ill than would have been the case in the past:

People who 20 years ago would be in a community hospital are now at home and people who 20 years ago might have been in an acute hospital are now probably in the community hospital.

KI-S03

It was suggested that if more people can be looked after in their own homes in the future there would be a reduced need for community hospital inpatient facilities, which may result in a reduction in the number of inpatient beds provided (KI-S04). There has been a small decrease in the number of community hospital beds in recent years, from 1473 in 2010 to 1328 in 2013,170 although it is unclear whether or not this was driven by a shift to home care. Key informants suggested that community hospitals might change their focus to provide increased support through outpatient services, investigatory services and ambulatory care (KI-S01, K-S0I4). Advances in technologies have increased the scope of procedures and investigations that can be performed and conducted in local facilities,13 with additional support from specialists through telemedicine (KI-S01).

Context and background of a health centre

The health centre was established in its current form in the late 2000s in the context of a merger of three municipalities into a single one (Municipality A). In 2012, the organisation and management of social care and health-care services in the municipality was consolidated through the creation of a ‘co-operation area’, which also included a neighbouring municipality (Municipality B). As a consequence, the health centre provides social and health-care services across two municipalities, with a catchment population of just under 30,000 residents. These mergers followed a major structural reform of the municipal system in Finland that sought to enhance co-operation between municipalities in order to ensure the availability of services (L169/2007). One of the key objectives of the reform was to integrate health and social services. 338 The organisation of the social and health services is agreed on in a co-operation agreement between the municipalities concerned.

According to interviews, the health centre served mainly an elderly population, with working age people being a minority among its service users. Table 24 presents the municipalities’ demographic information.

In 2010–12, the morbidity of the population in Municipality A was 107.3, and in Municipality B was 103.1. This compares with the Finnish average of 100. 340

The health centre serves a mixed rural and urban population; however, two-thirds of the population served live in a rural area. 341 Public transport services to the health centre are available, but were perceived to be poorer than the general standard in Finland (Kaikkonen et al. 342). The income level of the population served by the health centre is a bit lower than average level in Finland. 343

The Board of Social and Health Care (sosiaali-ja terveyslautakunta) is responsible for managing and organising the social and health services in the municipality co-operation area. The Director of Social and Health Services has financial responsibility for social and health-care services in the area. The range of social care and health-care services provided under the auspices of the board is presented in Box 12.

Overview of the health centre model

Balancing primary and specialised health care

The health centre is located on the same campus as a regional acute hospital, which is the nearest acute hospital and also part of one of the five university hospitals in Finland. The regional hospital provides specialised somatic and psychiatric services (Table 28).

The close proximity of the regional hospital was seen to facilitate co-operation with the health centre and access to specialised care services. According to interviews, this close co-operation made it easier for GPs and specialists to provide joint consultations and seek each other’s advice. Health centre GPs may also consult specialists at the nearby university hospital in writing and by telephone. Nurses in the health centre and the regional hospital were also reported to co-operate and consult with each other on service user care. In addition, the health centres’ rehabilitation function has also been strengthened by increased physiotherapy services on wards, especially since the rehabilitation ward was established and physiotherapy service was merged with the regional hospital.

However, according to interviewees, the physical proximity of the regional hospital also created confusion for service users, which has at times affected service provision. Municipality residents have been confused concerning the lines of division between the two organisational structures and the range of the services they offer, at times meaning that they are not aware of whether they are using health centre or regional hospital services. The physical proximity has also caused confusion around which hospital service users belong to, leading to the unnecessary use of specialised services.

One interviewee also highlighted that the proximity of the regional hospital had increased the use of specialist care services overall:

Traditionally, we have been using specialised medical care more than other municipalities have, simply because they are so readily available in the regional hospital. This closeness definitely has an impact.

S10

The interviewee perceived this situation as a conflicting one, within the context of municipalities’ purchase of health services:

[The] regional hospital is politically, concerning livelihood, a very important institution for us, but it has been threatened for years. Municipalities, which use services, have heavily reduced the purchase of these services, so the use of the regional hospital has been upheld, it has been very expensive for the municipalities.

S10

Indeed, in a difficult economic context, the health centre was seen to be somewhat deprioritised in terms of resource allocation in favour of the regional hospital, as specialised care has traditionally been most emphasised in the municipality under study. This was highlighted in interviews: ‘The regional hospital has always been powerful, so primary health care has been cast aside a little’ (S1).

Study participants reported pressure and demands for developing and improving service provision in the health centre using existing or even fewer resources. However, financial scarcity also acted as a driver for development:

If we have a limited amount of money and that is what we have to work with to provide the best possible care for people, the financial situation may be a motivator for finding new ways of doing things.

S4

Even if interview participants found that providing primary health-care services was expensive for the municipality, they emphasised the role of the health centre as a gatekeeper to specialised care, as well as the long-term savings that early intervention and prevention can offer. Participants noted that increasing attention had been paid to the health centre’s role in health promotion and disease prevention and that there was an expectation that this would eventually reduce the need for specialised care. The health centre was also seen as an essential local service for the residents of the municipality, and the proximity of the services was perceived to reduce costs. In addition, the provision of some specialised care in the health centre offered cost savings for acute services as well:

The most important thing is probably the fact that we can treat service users who were previously always referred to specialised medical care. In this way, service users receive care in their own locality. On the other hand, it has to do with financial matters as well I think . . . we also have our own special expertise, which may sometimes be quite advanced.

S6

The care provided by the health centre was considered to be more holistic than specialised care offered in the acute hospitals, as one interviewee expressed: ‘Specialised medical care only treats the hand, for example, whereas in primary health care, the person seeking care is seen more holistically’ (S5). Furthermore, the small size of the organisation and the familiarity of staff involved were perceived to enable more organisational flexibility and collaboration between teams, as one of the interviewees described: ‘Smaller-scale operations mean smaller personnel, which enables everyone to know each other and work flexibly together and not stick to your own duties only’ (S8).

Focus on service user independence

In the interviews, the objective of the health centre was defined as providing service users with comprehensive care and supporting them in their daily life, helping them to regain mobility as quickly as possible and supporting them in their ability to function:

Coping in the daily activities should be the starting point of development, whether we are talking about inpatient care or care in general practice. In any case, we should keep in mind what the needs of the person in question are.

S5

There was also an emphasis on person-centred service provision: ‘We should implement the idea that instead of the patient running from one desk to another, the physicians would go to the patient’ (S9).

When the co-operation area was established, the integration of social and health-care services was an interest shared by both parties. The integration resulted in the development of a shared culture for social and health-care services, and the operations adhere to shared objectives. With integration, care can be more easily brought to where the service users are:

Concrete manifestations of the co-operation between the social and health-care services include the ‘one-bed principle’ we have been applying for approximately 2 years. The principle means that if a patient already has a bed somewhere, then why not bring care there. In an acute situation where the patient requires medical care 24 hours per day, that bed is in the health centre hospital. Otherwise, care can be provided at home.

S7

Interviews indicated that the idea of the one-bed principle was based on user-centeredness, giving users the option to stay at home instead of in a hospital ward and for services to be brought to them. It required multiprofessional co-operation and the staff’s wide-ranging competence. At the time of data collection, services that could be brought to the users’ homes under this principle included intravenous treatments such as infusions.

There was a perception among interviewees that the integration of social and health-care services in this way had also helped to distribute health-care expenses more evenly. One reason for this was that service users could be transferred from a ward to their sheltered accommodation, where less expensive treatment could be continued. In addition, with this approach beds in the wards were freed up for service users to be transferred from specialised care for post-acute treatment.

Clinical skills and staff experience in the health centre setting

The health centre staff felt a sense of togetherness with their colleagues and in interviews were conscious of the importance of their work, as one interviewee indicated:

Now that we’ve been under the co-operation area for a while, we’re starting to feel like we belong. Achieving this team spirit just takes some time; to feel that this is our thing, our hospital, and we are really good at what we do here.

S2

Staff described the homeliness and innovation-friendly environment of the health centre, in which they felt comfortable. Interviews also indicated that there was high long-term employment and low numbers of sick leave days among the staff. In 2015, short-term sick leave days decreased significantly according to one interviewee. They also said that job rotation among staff members worked well and that colleagues were generally ready to cover shifts in the case of sudden sickness absence. All this may reflect a kind of loyalty to the health centre on the part of staff and strong job satisfaction.

The health centre as a whole was credited in interviews as being a pleasant place to work. Senior staff attributed this to the development opportunities offered to staff as well as openness to innovation:

We have not had a shortage of physicians since the joint municipal authority was dissolved and the co-operation area was launched. We have a good reputation as an employer, which is probably because we offer development opportunities and require that the personnel develop themselves. Employee input is also welcome, and personnel are given fairly free hands to innovate. This culture has attracted to our hospital acclaimed senior physicians, and younger doctors like to come learn from them.

S7

There was a perception among senior managers that staff in the health centre understood their role in contributing to successful recruitment:

We realise that unless we want to be understaffed in the future, every one of us must do our part to ensure successful recruitment. After all, people value the way they are treated far more than the money they earn.

S6

However, it remains unclear, from the data collected, whether or not this assumption was shared among all staff.

On the other hand, recruiting staff to the satellite clinics was difficult as it required lone working without the support of a larger working community. This was especially challenging in situations in which more junior GPs need support from senior GPs. Recruitment difficulties made the system vulnerable, for example in the case of sickness absence. For this reason, interviews indicated concern that the quality and availability of services could not be guaranteed in the satellite clinics.

In order to maintain and develop staff skills, an annual staff training plan was implemented for each operating area across the entire organisation. Training is provided within the hospital as well as in co-operation with the university hospitals. It was reported that requirements for staff competence have increased because of the expanding service provision of the health centre. It is providing increasingly comprehensive and advanced care, since the treatment periods at the health centre are becoming shorter and some specialised care is being provided. Training needs have also increased owing to the launch of new operations, such as mental health services. This was caused by functional changes in the hospital district when a psychiatric hospital was closed in 2015.

Staff pharmaceutical competence was maintained by regional pharmaceutical treatment training, which is mandatory for the entire nursing staff. The training enabled nurses to administer IV hydration in home environments, for example, supporting the ‘one bed’ principle.

The comprehensive and more specialised care offered in the health centre has also created a need to more clearly delineate staff roles. Practice nurses worked increasingly in home care and assisted living services, with registered nurses remaining at the health centre. This enabled registered nurses to co-ordinate service users’ care during the whole treatment period.

Furthermore, all staff (except for dentists) were required to participate in job rotation across different units of the health centre in order to maintain a wide range of skills. Job rotation was carried out in accordance with the individual’s needs and wishes, enabling them to align their choices with professional interests. In this way, skills-building work in the health centre aimed to equip staff to work in areas to which they were best suited; one interviewee described the principle as ‘[the] right people to [the] right place’ (S5).

The health centre also functions as a training site for GPs specialise in palliative care. This kind of specialism among GPs is rare in Finnish primary health care:

It’s one of the ‘crown jewels’. After all, we are one of the few health centres that are authorised to provide physicians with training leading to a special qualification in palliative care. Of course, there are such units at university hospitals and central hospitals, but these kinds of health centres and wards with that kind of training authorisation are few and far between in Finland.

S4

This function has increased co-operation with the relevant unit in the university hospital and the local hospice. It has also motivated health centre GPs to participate in training and increase palliative care competence in the health centre.

The importance of leadership training for supervisors was also emphasised in the interviews, in order to facilitate innovation and change management:

I believe that supervisors also need training in modern leadership. A solution-oriented way of thinking will help them implement changes in the organisation and provides them with leadership skills.

S5

Co-location and integration of services: benefits and challenges

Community engagement and demand

In Finland, a health centre is seen as a facility to which people go only to avail themselves of its services. There are no established voluntary work activities in Finnish health services. According to interviews, for municipality residents primary care was similar throughout Finland and the idea of a local identity for local services was not familiar:

Somehow I think that primary health care in Finland is quite a bit similar everywhere, if you go any municipality I think it has not very much local identity or at least I have not experienced it.

S1

However, it remained unclear whether the health centre studied did have a specific local identity for the surrounding community, as no service users were interviewed for this study.

The residents’ need for health services is assessed based on demand. This was demonstrated by the length of waiting lists, the quantity of referrals, as well as the service need observed by staff. Service user feedback was collected systematically and assessed, and used to inform service development. One example of a change implemented as a result of service user feedback was the introduction of a noticeboard in the waiting area of the emergency room, entitled ‘What am I waiting for?’. The noticeboard explained the operational model of the emergency room, as service users had complained about long and varying waiting times.

According to interviews, feedback received indicated that service users were relatively satisfied with the services at the health centre. Although the volume of feedback was not high, according to one participant this may indicate a positive, rather than a negative attitude:

But in practice, we receive very little verbal feedback. When the emergency service started here at the beginning of the year, it was a big change and resulted in a high volume of verbal feedback during the first month. We reviewed the feedback to determine what should be changed and implemented the changes. I can’t recall if I have received any feedback since March, and in practice, the feedback that comes to my level is usually major outbursts. The feedback volume is so low that I take it as a sign that people are happy with the services. If they weren’t, they would let me know.

S7

Negative media messages around primary health care, with little attention paid to the role of health centres, reportedly impacted on the residents’ perception of care provided in the health centre. However, systematic collection of feedback, information sharing and a service user-centred approach to care had had a positive effect on attitudes:

One patient in the specialised care hospital informed that she doesn’t want to go to the health centre, she demanded to be admitted to the regional hospital, but then the care provided by the health centre was described to her and finally she agreed to be admitted to the health centre. A couple of days was difficult for her but in the end of the treatment period she was very satisfied with the care and explained that she could have never imagined that the health centre can offer this good care.

S6

Future and the sustainability of the health centre

Interviewees supposed that the role of health centres would be reinforced in the future, and that this was in line with the apparent aims of the planned social and health service reforms in Finland:

I think there will be a lot of development for the better in the strengthening of primary health care. Also, although the social and health service areas [created under the reforms] will be large, the basic idea there, too, will be to strengthen primary health care.

S1

Maximising the care that can be delivered in health centres will also be made possible by increased specialised expertise in health centres in the future. Participants were optimistic about the potential for reforms to facilitate improved holistic, integrated and service user-centred care:

Let’s talk about the term ‘well-being campus’ that we’re trying to introduce. I think it would be great if our special fields would result in bringing down the silos and blocks between wards. After all, in practice, the majority of the patients are elderly, who would benefit more from an evaluation carried out at the general practice than a narrow assessment by an internist or a surgeon. Perhaps in the future we could use these special fields increasingly as consultants: our ward would be more of a general ward and specialists would come see patients here and we would discuss the situation together.

S4

In addition, the importance of the health centre’s future role as a preventative and health-promoting facility was highlighted:

We should find elements that help individuals take responsibility for the management of their own life and health. This would reduce the need for health-care services. Thus, prevention is important but new operating models should also be supported. The task of the health centre is also to treat illnesses at as early a stage as possible to prevent complications or the worsening of the condition, which would result in high costs.

S7

Advances in technology were also expected to help to allocate scarce resources to the most crucial areas of service user care and bring the services closer to the service user:

Laboratory tests even carried out by service users themselves and the results of the tests will be sent to their mobile phone.

S7

Context and background of the community hospital

The community hospital in question opened in a small village in a remote, hilly area of Italy in January 1996. Initially an eight-bed facility, it was established following the formal closure of the local acute hospital, by two municipalities belonging to the same valley. With funding from the local district, the former acute hospital premises were converted to house the new GP-run facility, which was considered an alternative response to the health-care needs of people living in the area.

After its establishment in the 18th century thanks to a bequest, the acute hospital had gradually increased in size and with regional funding became the main health centre of the valley in the 1970s. Since 1980, however, the number of wards was progressively reduced by the region from four to one, leaving only a general medicine ward with 20 beds. The hospital finally stopped functioning in 1993, after national registration349 required the closure (or conversion) of hospitals with < 120 beds. After 3 years of popular protests, the community hospital replaced the former acute hospital. In 2007, it was incorporated into a larger Casa della Salute (‘house of health’), with outpatient clinics on the first floor and inpatient beds on the third. Initially opened with eight beds in 1996, the number of beds in the community hospital had increased to 12 by 2001, and increased again to 18 in 2007 (Figure 22).

FIGURE 22.

Historical timeline.

Primary care in Italy has been traditionally provided by self-employed and independent physicians, GPs and paediatricians, under contract and paid according to their patient list, adjusted for age. In recent years, this independent practice model has been progressively modified and GPs are being encouraged to participate in collaborative arrangements to share practice space and resources with paediatricians and other professionals. 350,351 In this context, when faced with acute hospital closure, GPs in the two municipalities felt that working in a collaborative, interdisciplinary manner in the community hospital was the only option to avoid the complete removal of hospital beds in their area. The facility was established shortly after a similar but smaller hospital was introduced in a nearby village.

In 1996 our case study represented the first national experience of its type, in terms of the degree of local interdisciplinary team working, complex medical association and multiprofessional integration. 345 As never before, GPs jointly decided to work in synergy with other doctors and health professionals while maintaining their own patients through an outpatient service. This interdisciplinary collaboration allowed the community hospital to deliver a range of services in outpatient care, limiting the use of outside resources. The community hospital model established in 1996 also represented an innovation in the sense that clinical management, previously under the charge of hospital doctors, was awarded to GPs and out-of-hours care, to dedicated care continuity service physicians. Moreover, legislative references at the time of establishment were very limited (community hospitals were first cited in the National Health Plan 2006–8) and provided little formal guidance to the municipalities in their endeavour. Indeed, as one participant stated, ‘GPs had to use a huge amount of imagination’ (GP2), and to refer to international experience in order to develop the service model.

Following its establishment, the five GPs operating in the two municipalities agreed to conduct a 3-year experimental phase. In 1998, the local district’s evaluation of the testing phase produced positive results: types of admissions to the community hospital were comparable, at 92%, to those of the former hospital’s general medicine ward during its last year of operation. Thus, ‘the ex-general medicine ward and the community hospital were treating the same types of medical conditions, but with relevant cost savings in the latter case as it was entirely run by GPs’ (GP2).

In official terms, the facility was recognised as a ‘community hospital’ by its region only in May 2015, and before that operated for 20 years as a ‘country hospital’. Like the smaller hospital nearby (still categorised as a ‘country hospital’), from 1996 to 2015 the case-study hospital was entirely managed and financed at the municipality and district levels (from here onwards, when referring to the case study, we use the term ‘community hospital’, despite the fact that prior to May 2015 the facility was officially termed a ‘country hospital’). Although the services provided and the staff involved in the community hospital remained the same, accreditation as a ‘community hospital’ meant that the facility would now be financed by the region on the basis of a number of variables reported into a tailored database, the Regional Community Hospitals Information System (SIRCO). In addition to basic clinical, diagnostic and intervention data, the system focuses on measurements of patients’ conditions and independence, nursing care and social issues (D1, S1). These data do not serve as performance indicators, but rather as elements that characterise the admissions and determine the level of reimbursement by the region.

At the time of data collection, the community hospital was going through a phase of change triggered by the region’s recognition. The important changes to the financing and service model that this recategorisation entailed are detailed in Table 30.

A brief profile of the population served by the community hospital is presented in Table 31. The valley, encompassing an area of around 126 km², includes two villages (Village 1 and Village 2) that are about 10 kilometres apart. Its inhabitants include many elderly people who frequently live in an isolated manner, as suggested by the low population density. This is particularly evident in the smaller (and more peripheral) Village 2 compared with the Village 1, where the case study is located. The hospital mostly served the elderly population, and also accepts medically stable patients for short-term rehabilitation, social care assessment and palliative care. Over the years, ‘the community hospital became a landmark for the whole valley, providing locality-based care for patients without age limits, but predominantly older people’ (GP2).

Overview of the community hospital model

Currently located in a four-floor house of health, the community hospital is surrounded by gardens and a courtyard, with space for ambulance parking. An external lift facilitates access to the building for disabled people. The ground floor is dedicated to GP clinics and rehabilitation services, including a physiatrist and physiotherapist. The first floor hosts the outpatient clinics of specialist doctors (e.g. cardiologist) who also act as consultants for the community hospital on the GPs’ request. On the second floor, there are meeting rooms, offices and communal spaces. At the time of data collection, the community hospital ward consisted of 16 beds in total and occupied the entire third floor. It was composed of a single corridor with eight double rooms, a doctors’ hall and a space for intravenous therapy.

However, the building hosting the community hospital was reported to present some structural deficiencies that posed problems for patient accessibility. In particular, the exterior elevator was not spacious enough to carry patients on a stretcher, whereas the inner one was rendered difficult to access by the three steps that led to it (S1).

Staffing and services available in the house of health (encompassing the community hospital) are listed in Box 13.

Five GPs work collaboratively in the hospital and primarily manage patients belonging to their own list in the outpatient service. They dedicate 5 hours per week to covering the community hospital ward, divided into two weekly sessions of 2 hours each, as well as a weekly 1-hour GP meeting. The rest of their working time is spent at the outpatient service or in the community (e.g. home and nursing home visits). Meetings with nurses occur when requested by the nurse co-ordinator, and some participants considered additional meetings with other house of health professionals to be something that would be necessary in the future (S1, S2).

Medical and paramedical assistance is assured 24 hours a day by nurses, the care continuity service and GP shifts. Specialists who are not present in the house of health but whose expertise are often relevant to patient needs (i.e. urologists, orthopaedics, pulmonologists, diabetologists) can be found in the district acute hospitals or in the private system. In the event of complex medical issues, patients are transferred by GPs to the nearest acute hospitals or, in rare instances, a private specialist is invited to the community hospital by the patient (or the families) themselves (GP1). One ambulance with a nurse and a driver is ready to transport emergency cases to the nearest emergency service 24 hours a day.

The average bed occupancy is 12.6 (i.e. around 75%) and peaks during the winter period (for flu epidemics or COPD). Despite the number of beds increasing over time from eight to 16, bed occupancy is still below the target of almost 100% recently imposed by the region for community hospitals. Thus, some concerns were expressed by the staff about a possible cut in the number of beds in the near future (S1). As stated above, the community hospital admitted mainly elderly patients, most of whom were female. This is perhaps explained by demographic factors in the local area, as in 2014 there were almost five times more widows than widowers among the population aged over 70 years in the community hospital village (Village 1). 352 The key data on community hospital activity are reported in Table 32.

Capital costs to transform the previous small acute hospital into the country hospital were covered by the local district. Up until the time of data collection, the hospital’s operating budget was also covered by the district, with occasional assistance from the region for capital expenditure. However, with the recent change in status, the region was expected to take control of the hospital’s operational budget in the near future.

Community hospital’s role in the local health economy

According to interviews, the community hospital represented the ‘bridge’ between traditional acute hospitals and local services: ‘the community hospital was the ‘missing link’ between acute care hospital and community health services’ (GP2). It was seen as ‘the community service with the higher healthcare intensity’ (GP2). The GPs interviewed defined the community hospital as ‘the strongest response to the healthcare needs of citizens in the area’, underlining the local value of this model together with its relevance within the wider health economy. The community hospital responds to the needs of patients suffering from an exacerbation of chronic diseases, consequences of surgical interventions and terminal conditions that do not require intensive therapies or technology-intensive diagnostics. As mentioned above, patients are mainly elderly, with many in situations of isolation or living with inappropriate housing conditions for their health status (GP1). The conditions commonly treated in the hospital are listed in more detail in Table 33.

Although its primary aims are rehabilitation and safe discharge home, the community hospital has regularly served as a ‘hospice’ for terminally ill patients, especially those affected by cancer. Hospital admissions for oncological diseases were estimated at about 15% (GP2). Despite not being formally qualified as a hospice, in practice the decision to admit end-of-life patients to the community hospital has been justified by the distance to the closest hospice (16 km). Moreover, patients and their families prefer to be cared for close to home and ‘refuse to be treated in facilities expressly dedicated to palliative care, which are perceived as “anterooms of death”’. Finally, according to one interviewee, terminally ill patients not suffering from cancer had difficulty finding adequate care in traditional hospices, as they have only recently begun reorganising their services to manage non-cancer-related end-of-life conditions (D2).

Following the formal transition to community hospital status in May 2015 and the subsequent reduction of hospital stays to a maximum of 6 weeks imposed by the region (see Table 30), the issue of terminally ill patients is likely to become more pressing. Moreover, the new regulations will affect also non-terminal patients who have completed their treatment (or rehabilitation) at the community hospital but who are not yet ready for discharge. Indeed, community hospital beds were often used for extended periods while families were finding a suitable accommodation for the elderly. As one GP stated: ‘the community hospital was used as a temporary nursing home’ (GP1). A possible solution advocated by some of the interviewed stakeholders is ‘the enhancement of nursing home care service in order to allow most of cancer patients (or frail elderly) to end their lives in a family setting’ (D1).

Among community hospital admissions, staff interviewed distinguished two primary patient origins: (1) patients coming from acute hospitals for clinical stabilisation after an acute admission (mainly orthopaedic or cardiac surgery); and (2) patients coming from home (even under nursing care) who require hospitalisation owing to a worsening of their clinical condition (D2). Of 145 admissions in 2014, five patients came from the own district’s main acute hospital, 60 from another nearby acute hospital (belonging to another district), 77 from their own home and three from other facilities (e.g. nursing homes). As regards hospital discharges, 63.4% of patients returned home, 17.9% were readmitted to an acute hospital, 9.7% were admitted to a nursing home and 13 patients (9%) died in the community hospital.

The geographical location of the community hospital within the regional territory is peculiar. The two municipalities in its catchment belong to the same district, but the closest acute hospital (around 320 beds) is located in a neighbouring district. GPs therefore generally choose to refer to their own acute hospital (around 480 beds) for medical expertise not present locally (e.g. advanced diabetes mellitus care), even though it is further away. However, emergency cases are sent to the hospital of the neighbouring district owing to geographical proximity. For this reason, one interviewee stated that ‘the relationship with the district level is double and complex’ (GP1). In some cases, participants reported a lack of information exchange and co-ordination in patient management between the two acute hospitals (GP1, S2). In other cases, complaints related to long waiting time for patients admitted to the community hospital but referred to the district acute hospital for diagnostic tests emerged from the interviews (GP2).

As reported in Table 34, the distances between the two villages and the nearest hospitals are significant. The isolation is exacerbated by the lack of a rail connection (there is only a bus service). Travel distance represents a challenge in terms of time and fuel costs for families who care for patients admitted to one of the acute hospitals.

In this context, the community hospital has an important role in the wider health economy of its region. According to interviews, admissions to the community hospital can often prevent rehospitalisation to the acute hospitals or shorten the length of stay after acute admissions or surgery (GP2). Indeed, in October 2001 it was reported that the hospitalisation rate in the district had been reduced by 10% owing to the opening of the community hospital and the small country hospital operating nearby. According to interviews with GPs, emergency room visits were reduced, as patients could approach another doctor within the same health facility when their GPs were not on duty.

From an economic perspective, interviewees reported that the community hospital enables the containment of both health-care costs (as admissions are cheaper, in particular as a result of lower pay of GPs compared with medical specialists) and family expenses (owing to lower travel costs) (GP2, CR1).

Philosophy and aims of care in the community hospital setting

Interviewees reported a number of key clinical aims of the community hospital, which can be summarised as follows: (1) stabilising the overall clinical state and maintaining treatment outcomes achieved by the acute hospital; (2) preventing complications and recovering patient independence to allow a safe return home; (3) facilitating early discharge from the acute hospital and preventing rehospitalisation; (4) promoting access and engagement of family members and reducing their costs (i.e. travel expenses and productivity losses); and (5) enhancing the role of GPs and nurses by increasing their clinical and managerial responsibilities. 354

According to interviews, the philosophy of care in the community hospital relies on a ‘holistic approach’ to the patient, whereby the whole person, and not a single disease, is given full consideration (S1). In contrast to acute hospitals, the primary objectives of which are the diagnosis and treatment of severe acute conditions, the community hospital aims for the patient’s complete post-acute rehabilitation and social integration once discharged. A particular focus is put on psychological and social needs thanks to the co-operation of GPs with different professionals present in the house of health. As patients and doctors belong to the same community and they have already established a relationship in outpatient care ‘the local doctors’ deep knowledge of their patients is at the heart of this approach’ (GP1, GP2).

According to one of the GPs interviewed, the geographical location of the community hospital allows easier access to patients’ family, friends and the local community. Caregivers have a relevant role in treatment and rehabilitation and, for GPs, act as important source of health data, especially for patients with poor cognitive skills: ‘The involvement of patient’s relatives is particularly relevant in end-of-life situations and allows [us] to minimise the risk of legal issues and public complaints’ (GP1). The atmosphere was reported to be relaxed and friendly compared with the acute hospital and, at the time of data collection, there was greater tolerance for extending the length of stay and visiting hours. In referring to the community hospital, one of the doctors interviewed (GP1) spoke about ‘a facility rich in humanity and poor in technology’, pointing out that structural deficiencies (i.e. limited traditional equipment) are offset by a patient-friendly environment.

Clinical skills and staffing in a community hospital setting

Medical and nursing staff at the community hospital have been stable for several years and are mainly composed of professionals from the local area. Given that both patients and staff belong to the same community, they share knowledge and values shaped by similar local contexts and are also likely to know each other well. The staff interviewed confirmed that ‘the community hospital has a good working environment with friendly and established relations between different professionals’ (GP2, S1). With such stable staffing, the community hospital has not experienced any recruitment difficulties. In addition, the introduction of the community hospital has had positive effects on local employment, enhancing local skills and avoiding the depopulation of young people (CR1, CR2).

The community hospital also allowed for a welcome expansion of the traditional role for the GPs working there. GPs interviewed were grateful to have the opportunity to act as hospital doctors and to improve their clinical (and managerial) skills, a role that is not normally given to GPs in Italy (GP1).

Co-location and integration of services: challenges and benefits

The co-location of hospital and other health and social services within the larger house of health facilitated the integration of a number of care services. As mentioned, the availability of key medical specialties prevented the displacement of patients to acute hospitals. Moreover, the regular presence of a physiotherapist and other professionals allowed patients to spend the full period of post-acute rehabilitation at the community hospital. Co-operation among different professionals was enhanced by the introduction of the electronic case report forms, which enabled quicker information exchange (S1).

The GPs’ leadership style was reported to be participative (S2) and the importance of teamwork was emphasised in interviews. Indeed, in interview a member of GP management described himself as a ‘one among peers’ (GP1). Although in charge of liaising with district management, he did not consider himself to have enhanced power over other GPs either at the outpatient service or in the hospital (GP1). However, although the hospital GPs held weekly meetings, according to staff interviews, multidisciplinary team meetings were not regularly conducted. This was reported to be due to reluctance on the part of the GPs to share their leadership role with other staff. At the time of data collection, such meetings were expected to be periodically scheduled in the future. Moreover, according to some staff interviewed, future admissions and discharges to and from the community hospital would be agreed within a multidisciplinary team composed of GPs, nurses and other health professionals, in order to allow a more holistic perspective on patient care (S2, S1). Meetings with the district management occurred monthly to review economic data or to update clinical guidelines (GP1, S1).

In carrying out its core business of post-acute treatment and rehabilitation, the community hospital also depended on co-operation with acute hospitals which, while essential, also posed some challenges for patient care (GP1, GP2). With relatively modest medical equipment (comprising one electrocardiograph and two defibrillators along with common medical instruments), the community hospital relied upon the acute hospitals for access to a range of technologies. In terms of diagnostics, blood samples were sent daily to the acute hospital laboratory and the community hospital relied on the acute hospital for a range of other diagnostic tests (e.g. computed tomography). Although the community hospital lacked sufficient economies of scale to maintain these services, interviewees reported that reliance on the acute hospitals for such services at times led to unacceptable delays in patient care:

Their outsourcing inevitably creates some delays to the detriment of patients; the issue of waiting time will require a further evaluation in the near future.

GP2

According to interviews, the provision of appropriate care for the elderly demanded improved integration between health, psychosocial and social care services. Interviewees stated that with an ageing population, hospitals, and especially community hospitals, were increasingly required to meet social and psychological needs, which went beyond their traditional function (GP1, GP2, CR2). Social assistants, who may already propose hospital admissions for individuals living in poor conditions or without adequate family support, were expected to play a key role in this domain. Until May 2015, the social assistant’s collaboration with the community hospital was voluntary and informal. However, following regional recognition, their role was intended to become more formalised, with enhanced involvement in decision-making around patient admission and discharge. Historically, the social assistant has been in charge of assessing the needs of the elderly (through home visits or meetings with families), referral to services (e.g. meals at home, placement in nursing homes) and monitoring vulnerable people (through home visits or telephone calls) (S2).

The community hospital also works together with three nursing homes: one 28-bed public home adjacent to the community hospital; one 85-bed private home in Village 1 and a 39-bed private nursing home in Village 2 (CR2). Around 10% of patients discharged by the community hospital enter a nursing home. The facilities services (i.e. catering, cleaning) on which the community hospital relied were provided by a co-operative shared with the public nursing home.

Community ownership and engagement

Interviews indicated that the community hospital had a good reputation among the local inhabitants, which had been consolidated over time. Community representatives interviewed expressed a sense of ‘local ownership’ of the community hospital.

The strength of community feeling around local health provision and hospital beds was also indicated by the 3-year period of strong opposition among local inhabitants to the closure of the former acute hospital, which preceded the establishment of the community hospital. Indeed, in the early 1990s, delegations of citizens went to the regional authority’s offices to protest against the closure. The acute hospital, which had operated in the valley since the 18th century, had also been deeply rooted in the community.

The local community expressed a strong appreciation for community hospital provision. In the collective imagination ‘the [acute] hospital has never closed’ (GP2) and the community hospital was perceived as a direct continuation of the closed facility. Moreover, the community hospital staff is almost exclusively composed of local residents, which also contributed to the sense of local ownership and positive feeling. Indeed, the establishment of the community hospital allowed nurses and other staff members to return to live in their village after periods spent working elsewhere (S1, CR1).

Some participants outlined other key reasons for community satisfaction with the hospital, including (1) the proximity of family and friends during patient care and rehabilitation; (2) patient trust in the GPs thanks to their relationship established over time; (3) the sense of protection provided by the community hospital’s presence, which represents a beacon for chronic and post-acute care in the remote valley; and (4) the community hospital’s function in providing care for the elderly, which relieves the caring burden of the working population (CR1, CR2, GP2).

Volunteering activities in the valley are not particularly focused on health and social care, despite some initiatives that help to transport isolated elderly people to and from the hospital(s). No volunteers regularly work at the community hospital (S2, S1).

Future and sustainability of the community hospital

As described above, the community hospital is currently undergoing an important transitional phase. As with any change, the shift to different organisational and financing structures has posed some challenges for the community hospital. The passage from ‘country hospital’ to ‘community hospital’ will require GPs to give up some of their autonomy in favour of other professionals, as part of the multidisciplinary discharge meetings described above, for example. Moreover, the timely discharge of patients may be affected by the removal of the disincentive of a patient copayment after the 20th hospitalisation day.

As mentioned above, the community hospital’s role as an unofficial ‘hospice’, which it has had for the last 20 years, might also be subject to change. However, according to interviews, it is likely that the 6-week limit on length of stay will be extended for terminally ill patients. In addition, some interviewees indicated that the 6-week limit might be interpreted as average length of stay across admitted patients, rather than a strict deadline for each admission. Discussions with the region were ongoing at the time of data collection (GP2, D2).

With formal recognition as a ‘community hospital’ in May 2015, the region acknowledged the important part that the hospital has played in the regional health system in the recent past and is expected to play in the future.

Towards a common language: a push for regional standardisation and co-operation

Another challenge faced by community hospital staff as part of the transition to regional management has been the implementation of new management software, which is already widely used by acute hospitals. The system offered the possibility of connecting the community hospital with all the other district health facilities allowing them, for example to book laboratory tests directly. Despite attending training organised by the district, nurses and GPs reported that the programme does not fit the needs of the community hospital, as a facility that mainly deals with chronic and post-acute conditions. In particular, difficulties have been experienced in reporting complex therapeutic plans for chronic patients, with doses modulated according to patient characteristics. Moreover, one interviewee complained that the software was unsuitable for certain demands of clinical work and lacked sufficient technical support (GP1).

The ongoing transition from ‘country hospital’ to ‘community hospital’ was also reported to require the standardisation of clinical guidelines and practices among the case-study hospital and the other 28-bed community hospital in the district, located 40 kilometres away. These changes and moves for standardisation were intended to enhance community hospital integration with regional hospital provision, while still preserving the community hospital’s strong local identity and role (S1).

Co-operation between the community hospital and other regional health-care actors was also intended to be enhanced in other ways. Interviews also mentioned the potential for the development of a telemedicine service for the community hospital in the coming years. Telemedicine will allow doctors in the community hospital to receive the results of diagnostic tests (e.g. spirometry) in real time from the acute hospital and to share clinical information with isolated inhabitants in the valley while they are at home (GP1, S2).