INTRODUCTION
On 23 October 2014, NHS England Chief Executive Simon Stevens published a new 5-year plan for the NHS. Highlighting the challenges facing the NHS associated with an ageing population, the document argues:
‘The traditional divide between primary care, community services, and hospitals — largely unaltered since the birth of the NHS — is increasingly a barrier to the personalised and coordinated health services patients need.’1
It then goes on to discuss what models of care might look like under the plan. The focus is on integration and collaboration, eschewing further structural change but highlighting the potential of new service models which bring GPs together with a wide range of other providers, including community, social, and acute care services. Moving care closer to patients’ homes is highlighted, with vulnerable patients cared for proactively by multidisciplinary teams. None of this is new. GP fundholders pioneered better access to diagnostic tests and outreach by hospital consultants in the 1990s,2 while the 2000s brought Community Matrons, Virtual Wards, and Models of Case Management.3 However, integration between primary and community health services (CHS) has not been easy to achieve,4 and it is far from clear that such service models can, in fact, reduce costs.5
Against this background, an extensive review of existing literature was conducted to explore what factors should be taken into account in planning for primary care and CHS to work more effectively together. Starting with interdisciplinary healthcare teamworking (the micro-level), evidence was examined across all levels of the current care system to account for the diversity of the services.
At the meso level (that is, service organisation and delivery) this article focuses on whether services should be co-located and cover the same patient populations in order to work effectively together, and finally, at the macro level we explored structural aspects such as GP and CHS ownership and payment models that may influence joined-up working.
Our review suggests that there are ingredients for successful working to be found at the micro level, but that at the meso and macro levels evidence is hard to find.
A TALE OF TWO SERVICES
It is apparent that the current capacity of GPs and CHS to work together is largely determined by the history of the two services. Despite previous attempts to bring CHS and primary care together, such as the development of primary healthcare teams (PHCTs) in the 1980s/1990s and primary care trusts (PCTs) in 1997, the services continue to evolve separately. While much of this division stems from the inception of the NHS (when CHS and GP-provided services were separate in scope, funding, population coverage, and ownership), it is also rooted in different paths taken by waves of structural change in the NHS which have tended to reinforce barriers to joint working. For example, the reorganisation of community nursing into geographically-based neighbourhood teams (rather than attached to GPs’ practices) following the Cumberlege Report,6 and the more recent Transforming Community Services programme,7 which has seen CHS passed from PCT ownership to a variety of bodies, including private providers and third-sector bodies. The disconnection between general practice and CHS ownership therefore presents a key challenge in the ‘how to’ debate.
MULTIDISCIPLINARY TEAMS
At the micro level it is clear that enabling effective teamworking across service boundaries depends most on effective communication and contextual factors such as local geography and shared history. Shared IT and record systems can facilitate this, as can co-location of teams, although neither of these provides a complete solution. Characteristics suggested as being facilitative of team collaboration include good leadership and clear, agreed goals and objectives.8 Teams with a good internal ‘climate’ who work happily together have been shown to provide higher-quality care,9 and this in turn is likely to feed back to improve team climate, although evidence proving causation is lacking.
CO-LOCATION, PATIENT POPULATIONS: ARE THEY BETTER TOGETHER?
There is limited good empirical evidence across the meso and macro levels to provide conclusive examples as to how GPs and CHS can work effectively together. Research into the co-location of healthcare teams which cover the same patient populations suggests that this may have a facilitative effect,10,11 mainly by improving opportunities for communication. However, Ovretveit12 among others, makes the point that simply locating services together does not inevitably improve coordination of services. There is no clear evidence to indicate whether the current system by which GP practices and CHS care for different patient populations is better than caring for the same populations, with opinion divided between those who claim that organising nurses in geographically-based teams meets diverse population need better, and those who claim that providing nurse care based on GP-registered populations improves teamwork and coordination.
A variety of different service models have been piloted, such as federated practices and polyclinics, but few have been robustly evaluated and, as yet, there is no good evidence about the impact of such models. Combining practices into groups or federations that relate to a single community service ‘hub’ would seem to have the potential advantages of covering a shared geographical population and allowing a degree of co-location, but it cannot be assumed that such a model would be cost effective, necessarily reduce the need for hospital care, or even improve collaborative working.13
OWNERSHIP AND PAYMENT MODELS
Little is written about CHS ownership and payment models beyond what is known about the problems associated with the current block-contracting payment mechanism. There is no adequate evidence to link any one particular organisational form or ownership model with improved performance or service outcomes. What we do know is that the current diverse and fragmented nature of CHS results in a lack of data about CHS activity, which poses a significant problem, making it difficult to determine:
There is growing interest, however, in the alignment of GP and CHS payment models, with advocates arguing for the introduction of capitated contracts which incentivise CHS to produce high-quality outcomes. It is argued that such payment models would encourage innovation and joined-up working,14,15 although there is as yet no evidence to support this. Indeed, previous experience with capitated payment systems has tended to suggest that they risk incentivising under provision of services. Current proponents argue that this risk can be mitigated by paying for specified outcomes15 but this is as yet unproven. Finally, probably because of the difficulties associated with understanding and measuring activity, little is known about the ideal staffing model for CHS, either in terms of skill mix or numbers per head of population. One thing, however, would seem to be clear: caring for more patients in the community will require significantly more staff, and it is unknown whether there will be overall cost savings if the anticipated shift in care is achieved.
CONCLUSION
So, how can GPs and CHS work more effectively together to deliver Simon Stevens’ plan — is there a recipe for success? The short answer is no. The lack of good-quality research evidence available highlights the fact that while there are many contributory factors, there is no panacea. What does seem clear is that factors that facilitate effective communication are important, and that structural problems such as covering different patient populations and lack of a shared physical base can be overcome by promoting good relationships, and having sufficient communication channels. Allowing initiatives that improve community-based care to develop from the bottom up, building on successful local collaborations, rather than imposing a model from above, is another clear message. The interdependence of micro-level factors such as communication and shared local histories across the service (meso) and structural levels (macro) suggest that they are essential ingredients in most recipes for effective working.
However, it remains unclear exactly what an optimal funding model might look like.
Notes
Funding
This review was funded by the Department of Health Policy Research Programme as part of the research programme of the Policy Research Unit in Commissioning and the Healthcare System (PRUComm). The views expressed here are those of the authors, not the Department of Health.
Provenance
Freely submitted; externally peer reviewed.
Competing interests
The authors have declared no competing interests.
- © British Journal of General Practice 2015
