Universities and primary care organisations working together to recruit GPs: a qualitative evaluation of the Enfield clinical teaching fellow programme

Background General practice recruitment is in difficulty in the UK as many experienced GPs retire or reduce their commitment. The numbers of junior doctors choosing to specialise in the discipline is also falling, leading to primary care workforce issues particularly in 'hard to serve' areas. Aim To evaluate an academic service collaboration on GP recruitment between a primary care organisation (PCO), Enfield CCG, and a university, University College London (UCL). Design & setting Evaluation of an academic service collaboration in the Enfield CCG area of north east London. Method An action research method utilising qualitative methodology was used to evaluate a local service intervention, undertaken by the participants themselves. The qualitative data were analysed by one researcher but themes were agreed by the whole team. Enfield CCG, an NHS PCO, funded a collaboration with UCL to employ five GPs as clinical teaching fellows to work in Enfield, to increase patients' access, to provide input to CCG development projects, and to provide undergraduate medical student teaching in practice. Results Five teaching fellows were employed for ≤2 years and provided 18 266 extra appointments, engaged with development projects, and delivered local undergraduate teaching. The themes identified by stakeholders were the challenges of these organisations working together, recruiting GPs to an underserved area, and perceptions of the model’s value for money. Conclusion The evaluation showed that the collaboration of an NHS PCO and a higher education institution can work, and the prestige of being associated with a universty and clinical variety ensured GP recruitment in an area that had previously struggled. However, the project’s costs were high, which affected perceptions of its value.


Introduction
The incidence of complete Achilles tendon rupture is 18 per 100 000 patient-years 1 and is usually diagnosed clinically by GPs. The extent of clinical misdiagnosis is unknown in Norway, but may be high. 2 This is important as delayed treatment has unfavourable consequences. 1, 3 We report how a GP, with no clinical ultrasound experience, recorded images with a pocket-sized ultrasound device (PSUD) under supervision to confirm a complete Achilles tendon rupture. This could present a new indication for GP ultrasound.

Case report
A 36-year-old man experienced acute pain above the right heel accompanied by an audible snap while sprinting. He immediately had difficulty walking and 3 hours later consulted an on-call GP. Posterior ankle swelling with a tender depression 3 cm proximal to the calcaneum was found. Active plantar flexion against resistance was weak and Simmonds-Thompson test was 'partially positive' on applying a strong calf-squeeze. Based on these findings, calf muscle rupture was diagnosed as the Achilles tendon was thought to be intact. The patient was advised to elevate the foot and wait 2 weeks for improvement. Two days later a second GP, who was aware of a history of an audible snap, considered complete tendon rupture and reexamined the patient. Findings included an absent right heel raise due to weakness, minimal active plantar flexion against gravity and lying prone, significant right ankle swelling without bruising, and an altered angle of declination. Palpation elicited no ankle bony tenderness, yet a painful gap was identified 6 cm proximal from the calcaneal attachment, along the line of the Achilles tendon. Simmonds-Thompson's test was clearly positive. The positive Simmond's triad indicated a clinical diagnosis of complete rupture of the Achilles tendon. A 3.4-8 MHz linear array probe PSUD (VScanÔ dual probe, GE Healthcare), set at a depth of 3.5 cm, was used under the supervision of a rheumatologist experienced in ultrasound. The tendon was enlarged from 1 cm to 6 cm above the calcaneal insertion, where a clear gap was seen ( Figure 1). Two hours later a radiologist-performed ultrasound (LOGIQ E9Ô, GE Healthcare) and reported an enlarged distal tendon and a complete rupture at 5-6 cm from the calcaneal attachment, creating a 2.7 cm blood-filled gap ( Figure 2). Surgical exploration 8 days post-injury found a complete Achilles tendon rupture '5-10 cm above the ankle joint'.

Discussion
Tromsø Hospital serves a large area with a population of approximately 160 000. Between 2010-2014 an average of 21 patients per year were referred by their GP for suspected Achilles rupture.

Introduction
The incidence of complete Achilles tendon rupture is 18 per 100 000 patient-years 1 and is usually diagnosed clinically by GPs. The extent of clinical misdiagnosis is unknown in Norway, but may be high. 2 This is important as delayed treatment has unfavourable consequences. 1, 3 We report how a GP, with no clinical ultrasound experience, recorded images with a pocket-sized ultrasound device (PSUD) under supervision to confirm a complete Achilles tendon rupture. This could present a new indication for GP ultrasound.

Case report
A 36-year-old man experienced acute pain above the right heel accompanied by an audible snap while sprinting. He immediately had difficulty walking and 3 hours later consulted an on-call GP. Posterior ankle swelling with a tender depression 3 cm proximal to the calcaneum was found. Active plantar flexion against resistance was weak and Simmonds-Thompson test was 'partially positive' on applying a strong calf-squeeze. Based on these findings, calf muscle rupture was diagnosed as the Achilles tendon was thought to be intact. The patient was advised to elevate the foot and wait 2 weeks for improvement. Two days later a second GP, who was aware of a history of an audible snap, considered complete tendon rupture and reexamined the patient. Findings included an absent right heel raise due to weakness, minimal active plantar flexion against gravity and lying prone, significant right ankle swelling without bruising, and an altered angle of declination. Palpation elicited no ankle bony tenderness, yet a painful gap was identified 6 cm proximal from the calcaneal attachment, along the line of the Achilles tendon. Simmonds-Thompson's test was clearly positive. The positive Simmond's triad indicated a clinical diagnosis of complete rupture of the Achilles tendon. A 3.4-8 MHz linear array probe PSUD (VScanÔ dual probe, GE Healthcare), set at a depth of 3.5 cm, was used under the supervision of a rheumatologist experienced in ultrasound. The tendon was enlarged from 1 cm to 6 cm above the calcaneal insertion, where a clear gap was seen ( Figure 1). Two hours later a radiologist-performed ultrasound (LOGIQ E9Ô, GE Healthcare) and reported an enlarged distal tendon and a complete rupture at 5-6 cm from the calcaneal attachment, creating a 2.7 cm blood-filled gap ( Figure 2). Surgical exploration 8 days post-injury found a complete Achilles tendon rupture '5-10 cm above the ankle joint'.

Discussion
Tromsø Hospital serves a large area with a population of approximately 160 000. Between 2010-2014 an average of 21 patients per year were referred by their GP for suspected Achilles rupture.

Introduction
The incidence of complete Achilles tendon rupture is 18 per 100 000 patient-years 1 and is usually diagnosed clinically by GPs. The extent of clinical misdiagnosis is unknown in Norway, but may be high. 2 This is important as delayed treatment has unfavourable consequences. 1, 3 We report how a GP, with no clinical ultrasound experience, recorded images with a pocket-sized ultrasound device (PSUD) under supervision to confirm a complete Achilles tendon rupture. This could present a new indication for GP ultrasound.

Case report
A 36-year-old man experienced acute pain above the right heel accompanied by an audible snap while sprinting. He immediately had difficulty walking and 3 hours later consulted an on-call GP. Posterior ankle swelling with a tender depression 3 cm proximal to the calcaneum was found. Active plantar flexion against resistance was weak and Simmonds-Thompson test was 'partially positive' on applying a strong calf-squeeze. Based on these findings, calf muscle rupture was diagnosed as the Achilles tendon was thought to be intact. The patient was advised to elevate the foot and wait 2 weeks for improvement. Two days later a second GP, who was aware of a history of an audible snap, considered complete tendon rupture and reexamined the patient. Findings included an absent right heel raise due to weakness, minimal active plantar flexion against gravity and lying prone, significant right ankle swelling without bruising, and an altered angle of declination. Palpation elicited no ankle bony tenderness, yet a painful gap was identified 6 cm proximal from the calcaneal attachment, along the line of the Achilles tendon. Simmonds-Thompson's test was clearly positive. The positive Simmond's triad indicated a clinical diagnosis of complete rupture of the Achilles tendon. A 3.4-8 MHz linear array probe PSUD (VScanÔ dual probe, GE Healthcare), set at a depth of 3.5 cm, was used under the supervision of a rheumatologist experienced in ultrasound. The tendon was enlarged from 1 cm to 6 cm above the calcaneal insertion, where a clear gap was seen ( Figure 1). Two hours later a radiologist-performed ultrasound (LOGIQ E9Ô, GE Healthcare) and reported an enlarged distal tendon and a complete rupture at 5-6 cm from the calcaneal attachment, creating a 2.7 cm blood-filled gap ( Figure 2). Surgical exploration 8 days post-injury found a complete Achilles tendon rupture '5-10 cm above the ankle joint'.

Introduction
Last summer our small medical team visited the Calais 'Jungle'. Since that time much has changed and the camp is being demolished and by the time this article is read, it will probably be long gone. Some youngsters are finally being brought to the UK under the 'Dubs' amendment. However, once this camp is cleared it will not solve the ongoing flight of refugees from war torn areas: other camps are already appearing.

July 2016
A young Afghan man caught his finger on a sharp point while trying to cross a barbed wire fence. The finger was partially degloved. He attended the local hospital, where they placed a few sutures, but now, 2 weeks later, the skin is necrotic and the underlying tissue looks infected. He is in danger of losing his finger. A middle-aged Sudanese man has been having rigors and is generally unwell. He says it is similar to when he last had malaria.
A young Ukrainian woman complains of lower back pain and urinary frequency. The paths of these three people may never have crossed; yet here they are, denizens of the Calais Jungle. They turn up to a makeshift primary care 'clinic' that we set up in the heart of the unofficial refugee camp one weekend in July 2016.
With only basic medical supplies, we are immediately challenged by what we see. How can we arrange secondary care for the young Afghan in danger of losing his finger? We try to persuade him to return to the original local hospital, but he is reluctant. It was not a good experience for him the first time round.
With the other two patients, it is easier. They can attend the Salam clinic run by a local association during weekdays. Later, we receive word that malaria has been confirmed in our Sudanese patient.
More people arrive, presenting with scabies, rat bites, tinea, chest infections, and wheezing from inhaling smoke from fires lit to cook and keep warm in their tents at night. We examine a severely malnourished 2-year-old boy. We meet several of the camp's 600 unaccompanied children, at grave risk of sexual exploitation. We learn that there is inadequate safeguarding in place to protect them. A young Eritrean man comes in worried about his eye. He has sustained direct ocular trauma from a rubber bullet, and will never see normally again out of that eye. We see haematomas from police batons, and hear about children being exposed to tear gas again and again ( Figure 1).

The reality
These are no ordinary patients. They have travelled far from home to escape war, poverty, and misery. They have endured personal odysseys to get here, experienced untold hardships, and suffered unimaginable privations. Many have survived the loss of their families, torture, and rape. Their journeys over, for the moment at least, they must make their homes in the Calais Jungle. Their new shelters are in many cases mere tarpaulin covers, and their new beds just rugs on the ground. They own next to nothing. There is little for them to do, besides use their ingenuity to cross the English Channel in search of a better life. They are vulnerable to exploitation, crime, injury, and disease. Potentially violent clashes with local police, with other ethnic groups resident in the Jungle, or local far Introduction Recruitment to UK NHS general practice is an ongoing concern, particularly in inner city or more deprived communities. 1,2 The London Borough of Enfield is a socially-mixed area comprising both affluent and highly deprived suburbs. The borough historically has struggled to recruit and retain GPs. The most recent data (published in 2015) suggest that Enfield has 0.55 GP whole time equivalents (FTE) per 1000 population, which is below the London average of 0.57, and that there are '245 GP Performers across Enfield of which 25% (n = 61) are locums and 23% (n = 57) are more than 60 years.' 3 One of the questions which has arisen in this context is whether there is a potential role for universities' departments of general practice in helping to recruit GPs into 'hard to serve' areas. There are current national initiatives to address this issue in the UK, 4 and various local models have been piloted, including programmes linking GP posts to universities. 5-9 A systematic review of GP recruitment identifies several international programmes that combine an academic link with clinical service. 10,11 There are UK examples; from London in the 1990s, arising from the Tomlinson report and London Implementation Zone (LIZ), 12,13 which fostered projects such as the Inner City Lecturer scheme, 13 the London Implementation Zone Educational Initiative, 14,15 the London Academic Training Scheme (LATS), 16 and the South London Vocational Trainee Associates scheme. 17 Similar models have been employed in 'hard to serve' areas, in South Wales 9 and Durham. 5 More recently, there have been reports of university-linked localities, 18,19 and CCGs have been directly employing doctors. 20 These models have often been used by new GPs as a 'taster post' for a working in a specific community. 5 However, there is little evidence that these approaches are financially sustainable in the long-term, and there is mixed evidence on long-term GP recruitment. 10 English primary care structures have undergone substantial reorganisation since the LIZ and LATS programmes of the 1990s. 21 CCGs now have a statutory role in commissioning secondary care, but many CCGs are increasingly developing a contractor relationship (so-called 'fully delegated responsibility') with primary care. 22 These changes mean that CCGs and PCOs can now directly influence recruitment in primary care. Will these changes make these academic PCO models more sustainable?
Enfield NHS CCG is a PCO that commissions care. It allocated funding for a 2-year project and sought to collaborate with UCL in appointing four GP clinical teaching fellows (CTFs) to meet several service objectives.
The main aims for this initiative were to improve access by providing 17 000 extra GP appointments over 2 years; to deliver service improvements through research and re-design in priority service development area; and to raise the profile of Enfield for newly qualified GPs to settle in the long term. 23,24 The purpose of this evaluation is to examine the aims of the project and to assess success against its objectives. No specific funding was allocated to evaluate the project, so one of the implicit objectives was that the CTFs should evaluate the project themselves.

Method
This is an evaluation of a service innovation. 25,26 An action research approach was taken, 27 which is 'a collaborative process between researchers and people in the situation, a process of critical inquiry, a focus on social practice, and a deliberate process of reflective learning'. 28 Analysis of project documents and reports was used to capture stakeholders' views about meeting project objectives. A qualitative methodology was utilised, through interviews with key stakeholders. 29,30 Three of the CTFs acted as the action researchers and recorded interviews. The taped interaction between CTFs themselves and interviewees often generated rich discussion of the project, and so forms part of the data. The recordings were either transcribed verbatim, or annotated and summarised. The interviews were manually analysed using the thematic framework. 30 One author initially analysed the qualitative data using the following steps: familiarisation, identifying a thematic framework, indexing, charting, mapping, and interpretation. Throughout this process, transcripts were repeatedly re-read to focus on specific points and ideas, to verify the presence of the themes that are identified, and to ensure the context of the themes has been preserved. Themes were then discussed among the team and with key stakeholders. Divergent views were resolved by discussion. Drafts of the manuscript were shared with the CCG GP transformation subgroup to ensure, as the funder and a key stakeholder, that its perspective was captured.
The project is reported in line with COREQ guidelines (further information available from the author on request). 31

Recruitment
Participants were identified by the steering group selecting a purposive sample from each disciplinary group. If individuals did not respond to an email, a follow-up reminder was sent 2 weeks later. If they still did not respond, another individual from the same disciplinary group was approached. Participants received a study information leaflet and consent sheet to participate, to audio record, and to use quotes. The main ethical and methodological issue was that the researchers were also participants in the project. A grid was devised that would minimise potential conflicts or collusion through assigning researchers to interview stakeholders with whom they had had minimal role contact during the project. This pragmatic solution is discussed under study limitations.

Results
Several key documents were identified, including the CCG's own internal project final report. Other documents included project strategy and interim project management reports.
The CCG funded four full-time GP posts recruited for 2 years and employed by the university. GP CTFs were recruited by national advertisement, and with a joint interview and selection panel. The post attracted a strong field of applicants and four high calibre candidates were appointed. Their experience level ranged from newly qualified to 3 years post-training, and none had previously worked in the Enfield area. One of those appointed took maternity leave early on and did not subsequently return. Her funding was later used to appoint a fifth CTF, but this post was outside the scope of the collaboration. The CTFs' working week comprised five clinical GP sessions, two sessions for CCG service development, two undergraduate teaching sessions, and one session for professional development. Each CTF rotated to a different practice halfway through the 2-year programme in order to broaden their experience and share the additional workforce benefit that the CTF represented among different practices.
All practices in Enfield were invited to apply to host the CTF posts. Smaller, non-teaching practices were particularly encouraged to apply. Host practices were selected by application and eight were chosen. Service development projects were identified by the CCG and undergraduate teaching was introduced into 'teaching-naive' practices.
The CCG project internal summary report stated that the project delivered two of its main aims (unpublished data), but withheld judgment on third aim.
For the qualitative element of the evaluation, interviews were completed with 13 stakeholders. Data saturation was achieved.
All GPs interviewed are members of the CCG, unless identified as a 'university GP academic'. Those participants that are not board members (BM) are labelled as 'nonboard member (non-BM)' hereafter.

Objective 1: to improve access to primary care by providing additional patient appointments
There was a demonstrable increase in GP appointments: '... the clinical bit seems to have worked out and access has improved.' (GP CCG BM) The actual number of appointments delivered, calculated by the CCG, was 18 266; 7.4% in excess of the target of 17 000 (unpublished data).
Objective 2: to deliver service improvements through research and redesign The CTFs provided clinical input into service development projects (such as urgent care, diabetes, and end of life care) within the CCG with mixed results in terms of their abilty to embed in these teams and influence projects.

Objective 3: to raise the profile of Enfield for GPs to settle in the long term
Of the four postholders, two continued beyond the end of the post in Enfield practices and both have maintained a teaching commitment with UCL. The fifth directly-employed GP also continued to work in the area after the project. Two postholders did not complete the full duration of the programme; with one leaving to work abroad after 21 months, and the other due to family commitments after 11 months.
Undergraduate teaching was expanded in the locality and introduced into practices that were not previously involved. There were suggestions that this activity had an impact on the GPs, CTFs, and students in the practices in terms of their perceptions of Enfield as a place to work and study: 'Proud to say that [we] have had medical students into the practice.' (GP non-BM) '... more medical education will improve standards and the profile of the area.' (GP non-BM)

'When there is someone in the practice who is interested in teaching then it tends to 'rub off' on other colleagues' (GP non-BM).
Action research, in terms of reflective practice and problem-solving as a community of practice, occurred throughout the project, but was particularly evident around issues such as practice allocation, CCG project development, and implementation of teaching in surgeries.

Themes
From the qualitative data, four main themes emerged: 1. organisations working together; 2. recruiting GPs to underserved practices; 3. value for money; and 4. support and mentoring.

Organisations working together
Universities and PCOs are very different organisations with very different working methods and missions. A clear theme was the challenges for these organisations working together. This difficulty seemed to revolve around differing priorities within the shared project objectives, differing time scales, and differing cultures. The differing priorities revolved around service needs (delivering patient care and service redesign of clinical pathways for the PCO; teaching, CTF support, and -to a lesser extent -research for the university):

'... if the buildings are not fit for purpose you cannot have trainees [CTFs] in them.' (GP non-BM)
They were also serving high demand communities ('Deep End' practices). 32  There seemed to be reticence of some of these practices to apply to be hosts; possibly because they were already struggling with workload, so didn't have the resources to complete the application: Additionally, some potential practices didn't have enough consulting space to host another doctor or potential teaching space, which is the case for many inner city practices. 33 Larger, more organised, better resourced practices ultimately were more successful in their applications, and their link to GPs who were board members of the CCG gave the perception of inequity, leading to some local unease: '... number of GPs . . . hadn't realised that once they had stepped out of the steering group . . . they were conflicted, and wanted to actually apply to be hosts in their own right.' (CCG manager) '... it [the CTF programme] was advertised for non-teaching practices. It was allocated to teaching practices.' (GP non-BM) This led to representation from the Local Medical Committee (LMC), the representative body for local NHS GPs, 34 about the fairness of the process: 'Although assurances were made regarding the objectivity of the selection process, the LMC continued to have concerns -this led to scepticism locally with scheme.' (CCG manager)

Value for money
A recurring theme in the interviews and during the project was about the project's value for money. Funding four full-time GPs was a considerable financial burden for a CCG. The university ceded its usual full economic costing to make it affordable, which meant the project's finance was a source of concern for both organisations. However, several responders rated the project a success: 'I can only talk about our practice . . . I feel that we benefitted.' (GP non-BM) '... anecdotally, information from the practices about the positive effect within their practices also makes [me] feel it is a good investment.' (GP BM) '... from the teaching point of view, the post has been successful.' (University manager) However, success was more often viewed through a lens of 'value for money' by key stakeholders: '... the post has not been value for money for UCL considering the time and personnel that has been required to run this scheme . . . The true cost for UCL was not covered.' (University manager) 'It will be 'no' in value for money.' (CCG manager) '... hoping for a lot more from the post, but I feel that probably about 30% of these [ In order to justfy the project to both organisations, the objectives that were set were overly ambitious and, in hindsight, unrealistic: This raises the question whether this model of placing a CTF in a practice with multiple needs and no extra resource (beyond the additional clinical time that the CTF contributed, which was allocated to seeing patients) could ever make any changes. One responder felt the aim of 'Increasing capacity . . . has been achieved. However, this could have been achieved simply by employing locums.' (GP BM)

Support and mentoring
Due to the challenging nature of some of the practices and the wide scope of the project, the CTFs felt unsupported at times. The tight funding meant that there was little specific resource for mentoring and support: The fifth GP was appointed outside of the CTF collaboration towards the end of the project, perhaps to address some of the CCG's concerns about control and direction of the project.

Discussion Summary
This evaluation showed that universities can work with PCOs to promote recruitment of GPs in underserved areas. Using the brand or status of a university and making the role for the GP a varied one with service, teaching, and development can be attractive to GPs. However, several issues were identified that might be useful for those who may want to replicate such a scheme. Firstly, organisations need a stable collaboration to make such schemes work, with shared management and objectives from the outset. These objectives need to be clearly articulated and sufficiently robust that they can weather changing circumstances in both organisations. Unless very experienced practitioners (with consequent higher salary and project costs) are being employed, then there needs to be adequate mentorship (separate from line management). 35 There needs to be time to allow new appointees to settle into a post and to develop professional relationships before they can begin to initiate any kind of service change. Finally, any objectives set need to be limited in scope and realistic; if a GP is providing clinical service to an inner city practice, that is highly likely to spill over beyond any notional session time and will have an impact on their ability to undertake wider developmental roles. The consequences of these factors -lengthy posts, sustained support, and limited objectives -mean that they will be a significant financial burden on small PCO or similar organisation. That the collaboration unravelled and diverged towards the end of the project perhaps attests to the difficulty in sustaining such collaborations longer-term. However, the benefits of this model beyond patient contact are important: teaching medical students in practice is known to have an impact on career choice, 36 and on the choice of their working location; 37-39 teaching within a practice has a positive impact on practitioners 40 and their patients; 41 and there is value in developing links between PCOs and university departments of general practice. 21,42 Each of these attributes could probably be achieved more cheaply by employing locums to see more patients, incentivising GPs to work in an area, 43 incentivising practices to teach or train, or the direct funding of researchers by CCGs, 29 but narrow-focused models may not have the additive effect of this model.

Strengths and limitations
This approach enabled a low-cost means of evaluating a project that might otherwise go unevaluated, and generated useful data. The evaluation also captured a cross-disciplinary range of stakeholders' views.
There was no specific funding for this evaluation, so the researchers were the object of the evaluation. This will very likely have had an impact on what participants chose to disclose. While measures were taken to minimise this risk with interview schedules, there was evidence of participants who were known to be unhappy with the project not participating. Additionally, some participants were extremely cautious (they were perceived by the researchers as being 'very diplomatic') in what they would discuss. The summarised interviews may also have lost some of their context without the verbatim quotes.

Comparison with existing literature
The identified literature highlights many similar themes to this project. Specifically in relation to longer-term recruitment, the different models varied in their success. In the Durham model, it is suggested that 10-40% of postholders stay on in the locality; 5 75% of the LATS trainees stayed on London, 15 although this may just reflect a desire to stay in the capital. Retention in the south Wales scheme was high, at 61%. 9 More broadly, there were recognised benefits to health authorities, academic departments of general practice, and GPs working together. 15,20 Specifically, the CeMENT project reported evidence of the positive effects on GP from undergraduate teaching. 33 However, there is surprisingly little published evidence of the impact of these models on GP recruitment in inner city areas. 37 At an organisational level, there are difficulties with any health reorganisation, 18,23 and difficulties in establishing collaboration due to organisational tensions. 44 Within these collaborations, decisions and strategies will change over time, and will move away from their initial objectives, 13 but short term financial imperatives (usually cost savings) will usually win out over quality improvement activities. 42 This has been described as 'conflicting aims; interface issues; organizational problems; and leadership tensions'. 44 At the level of the specific interaction between PCOs and academic GP departments, the tensions identified here are echoed in other studies. 6 Specifically for this project, the long lead-time or bureaucratic delays reported here have been observed elsewhere. 15 One evaluation captures the governance issue of employment of GPs by institutions (such as universities) that are not providing GP services. 9 At the level of the individual GPs (be they fellows, LATS, or inner city lecturers), there is a clear need for mentoring during these projects, 5 which is usually provided by academic departments. 9,45 Practice-based mentoring caused issues in this study, but has been successful elsewhere. 5 However, peer support groups of these GPs appears to be an important element for project success. 5,9,45 This occurred spontaneously, locally, with this project but was not formalised. The ability to deliver local service development projects by fellows is highlighted, balanced against the very real issue of multiple competing priorities. 9,45 The need for continuity for the CTF in a single practice has been identified as necessary in order for them to become part of the practice team, 5 but in this project this was sacrificed to ensure equality of distribution of GPs to the practices.
For the host practices, the ability to apply and access such projects poses difficulties, 14 but this can be mitigated by a 'whole locality' approach. 15 Space -particularly in small, single-handed or inner city practices -often acts as a structural barrier to project participation. 33 For individual GPs within these practices or PCOs, there are often challenges with hosting, such as mentoring junior GPs, 5 engaging with practice changes, and the impact on the clinical service. 9 There are often financial conflicts of interest (around receiving free GPs) which need to be acknowledged and managed in a transparent manner. 46 Financial instability and unsustainability of these projects is widely reported, 6,16 with only the Welsh model, which had high level government support, seemingly able to provide a long-term mechanism to ensure survival. 9

Implications for research
This study shows that an academic institution can collaborate with a PCO to assist in recruitment of GPs to underserved areas, although there remain questions about financial sustainability. If mechanisms could be found to fund clinical time via the normal GP General Medical Services contract, this would substantially reduce project costs. The fact that these models have reappeared over the last 20-30 years suggest, while they are of interest to PCOs, they may not be financially or managerially sustainable.

Funding
The CTF posts were funded by the CCG; there was no specific funding for project evaluation. Enfield CCG supported the publication of this evaluation but have not influenced the findings.

Ethical approval
Health Research Authority confirmed this was a service evaluation and did not need formal ethical approval.

Provenance
Freely submitted; externally peer reviewed.