Abstract
Background GPs were a key driving force for the development of a network of community hospitals across England, and have provided medical cover for most of them. However, during the past decade there has been a significant shift, with the dominant trend appearing to be one of declining GP involvement.
Aim To explore how and why the role of GPs within community hospitals in England is changing.
Design and setting Qualitative study in a sample of nine diverse community hospitals in England.
Method Qualitative interviews with community hospital clinical staff.
Results In all, 20 interviews were conducted and two models of medical care observed: GPs employed by a practice and trust-employed doctors. Interviewees confirmed the trend towards declining GP involvement, with the factors driving change identified as being GP workload and recruitment challenges, a change from ‘step-up’ admissions from the community to ‘step-down’ admissions from acute hospitals, fewer local patients being admitted, increased medical acuity of patients admitted, increased burden of medical support required, and inadequate remuneration. The majority of doctors viewed community hospital work in a positive light, welcoming the opportunities for personal development and to acquire new clinical skills. GPs viewed community hospital work as an extension of primary care, adding to job satisfaction.
Conclusion Multiple factors have driven changes in the role of GP community hospital clinicians. The NHS needs to develop a focused strategy if GPs are to remain engaged with community hospital work.
INTRODUCTION
Practising local doctors have created, developed, and sustained rural local hospitals for more than 150 years. The first cottage hospital was developed in 1859 by Napper in Surrey,1,2 and by 1896 Burdett reported that there were already 240 such cottage hospitals in the UK.3
In 1920, Lord Dawson of Penn led a commission that set out a vision for primary health centres, enabling local doctors to work together.4 At this time, cottage hospitals were extending their services and facilities, and, over time, reflected Dawson’s design for local health care. Cottage hospitals became known as GP hospitals, with GPs providing medical services for inpatients, casualty, clinics, maternity, surgery, and anaesthetics. By the 1970s, Drs Rue and Bennett proposed that these hospitals become community hospitals, formally acknowledging them as an extension of primary care, serving the whole community.5
Cavenagh6 showed that, for England and Wales in the late 1970s, community hospital services equated to 20 district general hospitals, and that 16% of GPs were involved in their local hospitals. A subsequent UK-wide study in 2000 found that this had risen to 20%, a rate of one in five GPs.7 Features of community hospitals today are considered to be continuity of care, familiarity, accessibility, safety, and a community solidarity based on a sense of local ownership.8 Integrated care is also a feature of community hospitals,9–11 with GPs providing inpatient care as an extension of community-based primary care.
However, changes in the NHS have resulted in community hospitals being caught between a trend towards centralisation on the one hand, and for care closer to home on the other, which has contributed to a changing service mix, role, and function for community hospitals. The scoping and mapping work for a study of the profile, patient experience, and community value of community hospitals12 suggested that the role of GPs within community hospitals may be changing. Subsequent qualitative case studies confirmed a changing role for GPs within community hospitals, and identified several influential factors.
METHOD
This article is based on secondary analysis of data from a study that explored the profile, patient experience, community engagement, and value of community hospitals in England.12 This was a multimethod study that included: mapping, quantitative analysis of charity commission data on voluntary support for community hospitals, and qualitative analysis of 241 interviews and 22 focus groups with staff, patients, carers, volunteers, and community members drawn from nine case study community hospitals. The case study hospitals were selected to provide a diverse range of settings in terms of their age, size, range of services, geographical location, local deprivation levels, and levels of voluntary support. A detailed discussion of the methodology and full findings from the study as a whole are published elsewhere.12
How this fits in
GPs were instrumental in the establishment and staffing of community hospitals, and have viewed community hospital work as an extension of primary care. Over recent years, the level of GP involvement in inpatient care has declined. This study describes factors driving this change, including GP workload and remuneration, and admissions becoming less local, and more complex and time intensive. Community hospital work was acknowledged as being satisfying, and a stimulus for personal development. However, the NHS needs to develop a focused strategy if GPs are to remain engaged with community hospital work.
The authors focus here on secondary analysis of a specific subset of the data: 20 interviews from across the nine case study hospitals with eight GPs (including one joint GP and staff nurse interview), five specialty doctors or doctors employed by the trusts providing community hospital inpatient services, five nurse practitioners, and two GP practice managers. This sample comprised the main medical lead(s) for each of the case studies, apart from one where the lead doctor was unavailable for interview and the practice manager was interviewed instead.
The interviews were semi-structured, conducted face to face in 2016, and lasted approximately 45–75 minutes. The interviews focused on the main study questions (the profile, patient experience, community engagement, and value), with particular attention paid within this subset of interviews to the medical model operating in the hospital, how this had evolved over time, and personal experiences of providing it. Interviews were digitally recorded and transcribed verbatim, with all transcripts imported into the qualitative data analysis software package NVivo (version 11), which was used to aid data management and analysis.
An initial coding frame was developed after two members of the research team independently reviewed a selection of the transcriptions and discussed the emerging themes and codes together, and with the wider team. The emergent coding framework was then applied to all transcripts. A constant comparison approach to the analysis was used13 until saturation of the themes was achieved.14 Emerging findings were discussed with a number of the research participants at different points in the study to help ensure reliability of findings.
Although the analysis presented in this article focuses on this subset of interviews, it is grounded in the analysis of the wider data set created for the study as a whole, which was undertaken by the same research team.12
RESULTS
The role of GPs in community hospitals is changing. The dominant narrative to emerge from the nine case study hospitals involved in this study was one of a declining role for GPs. Whereas GPs had been heavily involved in all nine of the community hospitals in the past, not least through providing inpatient medical cover, the picture today is more complex. The authors identified two current models of medical care, distinguished by their employment arrangement.
GP employed by practice
Local general practices provided medical cover in some cases, with each GP partner taking part in the medical care of inpatients and providing a service for minor injury units. Alternatively, a GP with designated hours would provide medical cover for all the patients for their own practice, and also for other practices.
Doctor employed by a trust
These are doctors employed by the trust, with designated time for community hospital work; typically they are GP trained, and report to a consultant. They generally had no formal connection with local GP practices.
There was a disparity in the amount of dedicated doctor time between the two models, with less time generally for the GP medical model, without an immediately obvious relationship to the number of beds or acuity of the patients.
The general trend observed within the case studies was one of a shift from the first to the second model.
Factors driving change in the medical staffing model
Clinicians described a number of factors driving change.
GP workload and recruitment challenges
The increase in GP workload, rigours of the GP contract, and recruitment challenges were frequently cited as adding to the stress of fitting community hospital work into an already busy day. Having worked a full day in the practice, GPs described having to complete community hospital work in the evenings, with reports of GPs on the wards late into the night:
‘We couldn’t manage the workload. We have also a colleague, he tries to come late in the evening, but he was here till 12 am one day trying to admit somebody.’
(GP)
‘It is a significant addition to our current workload.’
(GP)
This was also commented on by nursing staff:
‘The GP didn’t come till 10.45 pm last night.’
(Staff nurse)
A change from step-up care to step-down care
Traditionally, GPs were able to admit ‘their’ patients directly from the community to the community hospital (step-up care), and would accept referrals of their own practice patients as well as other patients from an acute care hospital (step-down care). The findings show that the balance has shifted, with far fewer admissions likely to be GPs’ own patients:
‘I think when I first came here, a community hospital was meant to be for our local patients, so we could get anybody in, whoever we wanted.’
(GP)
However, all responders reported a change from step-up to step-down admissions to community hospitals. The main reasons cited were the pressure to admit patients from acute hospitals that were under strain, and the reduction in community hospital beds:
‘I would say step-downs outweigh the step-ups very, very significantly. If I’d have to guess, it would be at least 90% versus 10%, probably more.’
(GP)
‘The step-ups are very much reduced, and that’s due to our pressure in the acute, which feeds down to us.’
(Trust doctor)
‘We get a few [step-up admissions] but not many, so we see more of the step-down as they’re transferred from the acute hospital.’
(Advanced nurse practitioner)
‘They are only here because the [acute] hospital is full.’
(GP)
Decreasing localness of patients
The trend towards admitting more step-down and out-of-area patients, compounded by reductions in community hospital beds, resulted in frustration at not being able to admit local patients from the practice area:
‘Increasingly, however, our patients aren’t known to us because of the reduction in community beds. They’re often patients that have come from elsewhere … because of lack of continuity, you know. You’re meeting this patient afresh and they’ve got a list of medical problems and medication, but that’s not the same as knowing the patient.’
(GP)
‘We get a lot more people from outside the area … and there’s not that same link between GP practice and the patients.’
(Nurse practitioner)
‘What’s changed is that we no longer have control of who’s admitted.’
(GP)
Increased acuity of patients linked with attitude to risk
GPs, trust doctors, and nurses reported increased acuity of patients admitted to community hospital beds, and reflected on their attitude to risk in the light of this change:
‘Sometimes we have people who are stepped down too soon, are not as well sorted out as they could be, from a medical point of view.’
(Trust doctor)
‘Because the current hospitals are getting so busy, so they’re not being properly assessed, and they’re just pushed out from A&E and CDU [clinical decision unit].’
(GP)
‘We have had patients sent over on the transport, and we’ve blue lighted them straight back because there was an issue.’
(Staff nurse)
The attitude of doctors and trusts to managing risk was frequently mentioned as contributing to the decision to give up the role of community hospital GP clinician:
‘I think, historically, some of the doctors that worked here were more experienced in acute medical management, and their appetite for risk was higher. But I think the trust’s appetite for risk is relatively low. So we have this conflict between the trust saying, you know, “if you are unwell, you need to be managed in an acute hospital”, but then you also have people saying: “I don’t want to go to the acute hospital. I don’t want to be escalated.”’
(Trust doctor)
‘I think the degree of living with risk and things, which is a lot of our life, is probably — I think it looks a bit like that attitude has changed a bit, the balance of who is comfortable with what risk. Perhaps, quite rightly so. There’s a fear of litigation.’
(GP)
Increased burden of GP medical support
Linked to the decreased localness of patients admitted and the increased medical acuity was the perception of an increased burden of GP medical support. Because of a lack of a long-term relationship that a GP might have with their own practice patients and an absence of GP records, the amount of time required to admit and plan treatment was increased:
‘The GPs felt forced to take on patients from outside. It means that we spend significantly more time there than we used to 5–10 years ago.’
(GP)
‘You’ve got to meet a new person, get all the history, try and get some notes from the GP, which isn’t always so easy. Trying to get notes from the hospital isn’t very easy, and it’s double the amount of work than the patient you might know personally.’
(GP)
Inadequate remuneration
GP practices were paid for community work through a variety of local arrangements generally referred to as ‘the bed fund’. The payments were generally not felt to reflect the work performed, especially in the light of having to care for more complex cases not known personally to the GP:
‘GPs resigned the bed fund a year and a half ago … probably a mixture of the workload, the incredibly poor remuneration on the bed fund, and perception of clinical risk and adverse events.’
(Trust doctor)
Apart from the changing medical model of care, the narratives described mainly positive aspects to working in community hospitals.
A positive and enriching working environment
The majority of participants, GPs, and trust doctors viewed community hospitals in a positive light, and as vital local resources to be fiercely defended. For some, their whole professional life, and to some extent their personal life, had been deeply engaged with the community hospital:
‘I’ve had my own family involvement there, with my father being treated there post-operatively after his coronary arteries bypass grafting, and in dying there.’
(GP)
‘Oh, I’m deeply attached to E2 Hospital, yeah, deeply attached. I live here. I work here. I brought up my family here. My children have been to E2 Hospital.’
(Trust doctor)
The holistic care, chance to meet patients and their carers over a longer period of time, and the ‘low tech’ environment of community hospitals were all seen as positive features. The service to isolated rural populations, and the concept of offering a choice apart from referral to an acute hospital setting, were seen as benefits to the local community:
‘I think community hospitals provide us an insight into this thing, that 80-plus or 85-plus patient, look into him holistically in a complete sense.’
(Trust doctor)
‘I get to know their families and that kind of accelerates knowing community more.’
(GP)
Both GPs and trust doctors reported opportunities for personal development and increasing skills as a result of working in community hospitals, leading to job satisfaction:
‘Certainly, my skillset has changed. I am better at dealing with more acute medicines doing this job than I was when I was just kind of a more normal GP.’
(Trust doctor)
‘I like the role. I think it provides a lot of diversity … it’s a bit different. It also adds that little element of acute care onto it, so getting slightly sicker patients, a bit more complex. So it’s really good for my development, in terms of learning how to sort of manage these complex patients.’
(GP)
‘It’s been my professional life, and it’s been a very satisfying professional life as a GP and hospital doctor working in community hospital. The two have fortified each other, and [having] been completely browned off, fed up, and burnout in general practice, I come down here and be nourished.’
(Trust doctor)
Although doctors were generally positive about community hospitals, some did raise questions about the future of community hospital inpatient beds.
They questioned whether the beds were financially viable, and if they were the most appropriate way to look after older patients:
‘We don’t need a community hospital with beds. We need community resources to do the re-ablement and the rehab in people’s homes.’
(GP)
DISCUSSION
Summary
The role of GPs in community hospital inpatient care is changing, mostly declining. The reasons for this change are complex. The trend towards admitting patients from a wider area, resulting in a loss of personal continuity and a more demanding admission process, along with loss of ability to admit their ‘own’ patients, has led to disengagement. The complexity and acuity of patients was reported to have increased, necessitating more time on the ward, which is hard to fit in to an already busy day. Recruitment challenges in primary care, and a sense of not being valued, have also contributed to a change in the medical model.
The majority of doctors who were interviewed viewed community hospitals in a positive light, and valued the ability to provide holistic care closer to home, and integrated care between the acute hospital and home. For both GPs and trust doctors, community hospital work provided job satisfaction and opportunities for personal development, and to use and extend clinical skills. However, some questioned the future viability of community hospital inpatient beds.
Strengths and limitations
The strength of this study is the collection of in-depth qualitative data from clinical staff located in a varied sample of community hospitals in England. All doctors interviewed had training in general practice, and some had previously worked as GPs before becoming trust doctors, and hence could see the advantages and disadvantages of both models.
Weaknesses of the study were the relatively small sample size and that no specific enquiry was made by the researchers regarding specific training for community hospital work and future training needs.
Comparison with existing literature
The literature around GP involvement in community hospital inpatient care is sparse. A study from 1988 recorded that, on average, 16 GPs had admitting rights to each community hospital, and that three-quarters of GPs viewed community hospitals ‘very favourably’.15 A survey examining the situation in 2000 estimated that one in five GPs in the UK had admitting rights, with nearly 100% of inpatients being cared for by their own GP or a GP from the local practice.7 Observations from the Community Hospitals Association (CHA) indicate a reduced involvement of GPs in providing clinical care (CHA, personal communication, 2018); however, exactly how many GPs are currently involved in hospital inpatient care is not known.
The concept of community hospitals as extensions of primary care has been described before,16 with the importance of personal care by a patient’s own GP being emphasised in a Scottish study.17
A study examining community hospital care across a number of countries observed an increase in step-down care in Scottish community hospitals in response to pressure to free up beds in acute hospitals, and a reduction in beds for direct GP admissions.11 Likewise, difficulties in recruiting both nursing and medical staff to work in Scottish community hospitals were observed. The same study reported an increased complexity of medical cases observed in Italian community hospitals, and a decrease in ‘procedural GPs’ in Australian rural community hospitals, probably related to closure of smaller units.11
Implications for research and practice
The authors would argue that the reduction in GP involvement in community hospital inpatient care is a significant issue. Community hospital work fits well with the generalist skills of GPs, who are able to provide intermediate care for the local population and is line with NHS priorities. Patients place a high value on community hospital services, and have significant engagement with ‘their’ community hospital.12
GP involvement with community hospital care has long been an important aspect of this engagement, and reduction of GP involvement may reduce the strength of connection between the hospital and the local population. Rural populations are often poorly served by an increasingly centralised health service, with reduced access to services and increased travel demands. The opportunity to be cared for locally by a doctor known to the patient is highly valued by patients,12 and withdrawal of GP medical support weakens this aspect of care.
At a time when there is a move towards more care in the community and the integration of health and social care, it is of concern that there is widespread closure of community hospital beds18 and withdrawal of GPs from providing integrated care for their patients, both at home and as inpatients. Taken alongside rising pressure on acute-sector beds and rising GP workload, it would appear that community hospitals have a role to play in alleviating these pressures. The lack of a national strategy for English community hospitals is in contrast with other countries that recognise their value and actively plan to increase provision. In Norway19 and Italy20 government legislation allows local authorities to provide services that are alternatives to acute hospitals, and one of the models developed is the use of municipal acute care beds with GP medical care. In Scotland, a clear strategy for developing community hospitals was issued in 2006,21 and refreshed in 2012.22
This study describes two models of medical inpatient care, but it is unclear which provides the ‘best’ clinical care. Arguments were heard for both models: GPs enable continuity of care and could improve patient experience, whereas trust doctors reduce risk because they are on site more of the time, and provide more consistent care. Further research is needed to explore clinical outcomes and patient experience associated with both models.
A quantitative survey would assess the extent of GPs’ involvement and the requirements of community hospital doctors regarding training, peer support, and help in managing risk.
The NHS needs to develop a focused strategy if GPs are to remain engaged with community hospital inpatient care. There is currently scope to examine the association between clinical outcomes and the two models of medical care described.
Acknowledgments
The authors would like to thank all responders who contributed to this study through participating in interviews, focus groups, site visits, and responding to requests for information. Particular thanks to the matrons and ward sisters, community hospital clinicians, and League of Friends members who coordinated research activities, organised local reference groups, and participated in annual learning events.
Notes
Funding
Funding was received from the National Institute for Health Research — Health Services and Delivery Research. New research on community hospitals 12/177/13.
Ethical approval
The Wales Research Ethics Committee 6 approved this research study (reference: 16/WA/0021).
Provenance
Freely submitted; externally peer reviewed.
Competing interests
The authors have declared no competing interests.
Discuss this article
Contribute and read comments about this article: bjgp.org/letters
- Received July 4, 2018.
- Revision requested September 16, 2018.
- Accepted October 10, 2018.
- © British Journal of General Practice 2019
