Evaluation of Integrated Resource Framework Test Sites

5 TRIALLING NEW MODELS OF INTEGRATION

5.1 The IRF process was clearly defined by the two complementary elements of: (i) jointly mapping resources and activity across health and social care; and (ii) developing new protocols to allow resources to flow between NHS and Local Authority partners. Ayrshire & Arran, Lothian and Tayside all chose to develop new ways of working with a number of smaller pilot populations based around localities, care groups or disease programmes. These pilots ranged in size, focus and purpose. In contrast, at the end of 2010, Highland test site took a decision to bypass the process of trialling small scale pilots of change as a means to testing out the potential of integration. Instead NHS Highland and Highland Council made a commitment at the most senior level in both organisations to undertake integration and realignment on a much larger scale using a lead agency model (see chapter 6).

5.2 At an early stage in the IRF programme, the Scottish Government set milestones in the expectation that test site trials for resource realignment would have new financial mechanisms in place by April 2011. However, none of the test sites were actively running new integrated financial and governance arrangements by the time of the final evaluation in March 2012.¹⁵ Therefore, this evaluation is limited to analysis of the processes of setting up new protocols and does not comment on the impact of trialling resource realignment. This chapter reports on the degree of progress made in introducing new financial mechanisms and analyses the key challenges encountered.

Progress

Adding fresh impetus to existing partnership working

5.3 The IRF methodology was new and innovative but was introduced in the context of ongoing work to integrate services. This meant that test sites designed and implemented the IRF pilots in a range of ways reflecting local priorities and needs. This approach was reported as a positive opportunity for local partners to initiate new areas of joint working as well as developing existing areas. It built upon partnership relationships that were already in place as well as introducing new partners. However, this made it more difficult to draw a clear distinction between IRF pilots and other local partnership and national initiatives and policies. The following list exhibits the types of pilots that emerged:

• Pilots that were initiated by the IRF process and were resourced with additional IRF funding
• Local pilots that had been started prior to the IRF but were boosted with IRF funding and support
• Pilots that used IRF funding alongside other initiative funding for different policy priorities
• Pilots that were included within IRF learning and reporting structures but did not receive any IRF funding

5.4 During the course of the evaluation, the test sites repeatedly gave examples of other partnership working initiatives which were not labelled as IRF but which nonetheless developed or demonstrated learning from an IRF approach. The growing number of partnership initiatives incrementally reinforces a culture of integration where traditional boundaries can be challenged and changed.

"You could get rid of the 'You work for social work, I work for someone else' and instead say 'we are the dementia team, we work for these people and we work these hours and conditions'"

Using local knowledge of need and opportunity

5.5 The selection of IRF pilots reflected local partners' knowledge of areas where partnership working was already recognised as having potential. This was usually based on jointly recognised priorities and an accepted sense of opportunity for service review and improvement.

5.6 For example, NHS Lothian and City of Edinburgh Council (CEC) partners decided that the implementation of a new model of care within orthopaedic and stroke rehabilitation pathways had considerable potential to shift the balance of care from hospital to community settings. This decision followed on from the NHS Lothian Strategic Model of Care and Capacity Review for Older People in 2006, and was informed by benchmark information. A phased implementation of this approach was planned during 2010/11. This coincided with Lothian's bid to become one of the IRF test sites, making it logical to use this project as the IRF pilot for the CEC partnership area. The aim of the model was to shift the balance of care from hospital to community settings, with a related objective to enhance rehabilitation in hospital in order to increase the functional level of patients at their point of discharge.

5.7 Similarly, the decision to select NW Perthshire as one of the Tayside IRF test site pilots was influenced by a range of local factors. These included successful community engagement around reconfiguration of community hospital beds, the particular needs of a rural community, and an interest in extending successful models of joint working to deliver more integrated health and adult social care services to a wider integration agenda.

Focusing effort and adding detail

5.8 The pilots were used by some test sites to take a more detailed look at the types of information that could inform integration. They identified populations based on geographic boundaries, age groupings, care pathways and disease groupings. The more detailed work on costs and activity for these populations addressed some of the limitations of the much broader mapping exercise commented on in chapters 3 and 4. The pilots focussed on: patient level costs for an identified patient population; staff views on service provision for identified patient populations; and potential changes to service provision and anticipated outcome measures.

5.9 Some pilots initially defined a scope that was too large to undertake within the prescribed resources and timescales. For example, initially the work in Ayrshire and Arran attempted to look at young people with complex needs across North, South and East Ayrshire. However, in order to make progress the scope was revised to a more targeted piece of work within North Ayrshire.

5.10 The pilots that appeared to make quicker and more obvious progress took a more pragmatic approach and restricted the scope of their ambition to a manageable geography, population, number of service partners and impact. For example, the Lothian work on phased implementation of orthopaedic and stroke rehabilitation pathways involved a very specific patient population, namely orthopaedic patients in two wards at the Royal Victoria Hospital and stroke patients in one ward at the Astley Ainslie Hospital.

Encouraging leadership

5.11 The test sites put in place structures to organise the IRF process. Senior strategic partners were represented on steering groups and implementation groups with the intention that these would coordinate and drive the mapping and pilots. More than three quarters (77 percent) of service delivery staff who responded to an email survey reported that there was senior buy-in to the IRF from their organisation. However, this did not necessarily equate to leadership since only 56 percent of the same group of respondents reported that the IRF had visible leadership.¹⁶

5.12 The evaluation found evidence that leadership was more effective where senior managers were actively involved within both partner organisations. It was also reported that leadership could be provided by staff at different levels within each organisation who took a proactive approach to encouraging colleagues to consider new integrated ways of working. Positive leadership and trusting relationships were reported to be crucial to building momentum for change.

"'Barriers are really down to personalities - mutual understanding and relationship, irrespective of structure, is what makes a difference."

5.13 The experience reported in Highland (see chapter 6) indicates that political and executive level leadership is key to enabling full scale trialling and implementation of new financial mechanisms and governance arrangements. This was reported as necessary to overcome the inertia within different systems which implicitly acts to dampen the scale and rate of change.

Giving staff time to reflect on practice

5.14 Integrated working was reported to be happening in a 'bottom-up' manner. Front line staff in health and social care reported examples of how they work together - and want to work together more - for the benefit of service users. This was often reported to occur despite organisational structures and systems rather than because of them. The pilots re-enforced examples from other partnership initiatives in showing potential localised benefits of local staff being given time to reflect on practice and develop and lead service improvements. For example, in one of the Perth and Kinross workstreams, front line staff reviewed the discharge planning process, developed and integrated an efficiency improvement that was reported to have freed up a significant amount of social work time. Similarly, in East Ayrshire, the IRF enabled existing staff to take time to research current services for adults with complex needs and create recommendations for future improvements.

Introducing the philosophy and language of commissioning

5.15 The test sites reported that the IRF programme had been helpful in providing examples of models for trialling new financial mechanisms; in sharing experiences from elsewhere in the UK; and in publishing a literature review of evidence in this area.¹⁷ The test sites reported improved understanding amongst stakeholders of the terminology of commissioning and the possible financial mechanisms involved. For example, Ayrshire and Arran undertook a specific workstream to develop locally agreed governance and financial mechanisms to support local integrated working. This resulted in a document outlining the possible integrated governance arrangements.

5.16 Partners reported an increased awareness of the value of introducing some form of commissioning to link the cause and effect of spending decisions. It was reported that the IRF had helped to articulate the instability within the health and social care system where currently those making decisions on the place and type of care are not directly affected by the costs of their decisions. It was acknowledged that integration needs to legitimise the tension across the system so that those who make decisions have some form of direct feedback about the resulting cost implications.

"If a CHP wants to influence a hospital then it doesn't need to know how much each procedure costs or detail on a patient by patient level. What it needs to know is that if GP practices make changes that result in reduced admissions and reduced length of stay then that frees up a certain amount of cost that can be looked at by partners."

5.17 The evaluation found evidence across the test sites that joint strategic commissioning needs to be more explicitly implemented in the health and social care system to address the issues raised by improved information. The mapping data and the trialling of new ways of working raised the profile of variation, opportunity cost and potential improvements in efficiency. However, it was reported that integration needs both information and incentive to effect change. The IRF brought into sharper focus the difficulty of addressing these issues when there is no direct financial mechanism to link decisions concerning demand and supply across the whole of the health and social care system.

Challenges

Uncertainty and speculation

5.18 The IRF, the Change Fund and the Cabinet Secretary's announcement concerning the reform of CHPs, all share a common objective to deliver more integrated health and social care.^18,19 The consistent policy direction was reported to be helping to create a culture of integration. However, the policy direction outlining the extent of service changes emerged over time, and this was linked to a degree of uncertainty amongst local partners. During 2011, stakeholders reported increasing uncertainty caused by speculation about national proposals for changes to Community Health Partnerships which could involve greater integration of budgets and accountability. This led to uncertainty about the IRF and some stakeholders reported a degree of reticence about expending effort to trial new financial mechanisms until the national direction was clarified. The notable exception was the Highland test site where, in April 2012, NHS Highland and Highland Council introduced a Lead Agency model for Health and Social Care with delegated budgets for adult social care.

Scale of integration

5.19 The IRF pilots aimed to enable test sites to try out integrated financial mechanisms on a limited scale, with the expectation that these would produce a stronger evidence base which would encourage wider implementation. However, progress in the pilots was limited by the fact that some were small scale trials and therefore did not receive the long term commitment needed for partners to deliver much more than marginal change. However, as the evaluation developed, there were signs of progress, such as the consumption budget work in Tayside and the Lead Agency model in Highland, which continue to build momentum at a more significant level.

Transferring resources

5.20 As noted above, by the time of the final evaluation, the IRF test sites had not yet implemented new governance arrangements or financial mechanisms and there was a lack of evidence that the IRF had influenced a change in the use of resources. A quarter of service delivery stakeholders surveyed indicated that the IRF had helped clarify decision making but only five percent reported that it had enabled the flow of resources between health and social care.²⁰

5.21 Partners involved in the research cited examples from the IRF and other partnership initiatives which have made better use of resources in one part of the health and social care system to change the place or type of care for particular groups of patients. However, the evaluation did not find evidence of this sort of change leading to savings being released or resources being transferred across different parts of the systems. Analysis by the test sites indicates that the underlying issue is poor cross system demand management which back-fills any space created by new ways of working.

Addressing fixed costs

5.22 Fixed costs remain a big challenge for health and social care. The IRF has helped to support the case for shifting the balance of care by continually placing the idea of opportunity cost into discussions on the best use of resources. However, better mapping data and improved dialogue amongst partners have not yet evidenced an influence on decision making to disinvest resources which are tied up in staffing or buildings.

"The big barrier that is starting to be acknowledged by stakeholders is that the reality of shifting resources is being hampered by different definitions of fixed costs. If beds, consultants, wards and ward staff are all regarded as fixed costs, then the options for shifting the budgets associated with the improvements aspired to in shifting the balance of care are unlikely to be achieved."

5.23 The mapping data and the work undertaken by IRF pilots have started to identify areas where integrated working could create both improved ways of working and changes in the demand structure across the health and social care system. However, stakeholders reported that fixed costs form a barrier in shifting towards more integrated ways of working. Three main areas were identified: (i) a negative public perception of changes which lead to a reduction in physical resources such as hospital wards or day care centres; (ii) political commitments to protecting changes in staffing levels; (iii) an historical mindset which persists in inappropriately defining the majority of costs as fixed.

5.24 It was reported that the case for addressing fixed costs could be enhanced by:

• A better understanding of how cost and activity data for health and social care is linked with patient outcomes
• A better evidence base for the risks associated with changes in recognised care patterns which is also better shared and understood by all those involved (politicians, planners, professionals, patients and the public)
• Partnership structures which have better representation of all relevant stakeholders and also the power to change the use of resources across care settings and care pathways
• Referring to examples of success in reducing fixed costs where this is planned and managed over a defined period, for example the deinstitutionalisation of mental health care in the 1990s.

5.25 Decisions to disinvest in current staff and buildings are highly contentious and require: a stronger evidence base of the benefits of disinvestment; committed leadership from politicians and service Chief Executives that fixed costs can be redefined; greater public confidence in decision making bodies where fixed costs are to be redefined; and clearer communication of the motives and benefits of changing the composition of buildings and staff used to deliver services.

Balancing representation

5.26 Stakeholders involved in the process reported that it was much simpler to get representation on partnership projects from people who were involved in management than it was to get meaningful representation or engagement with groups such as GPs or hospital clinicians. One of the key challenges was the significant variation in geographic responsibilities of different professional or staff groups. Some examples of the complexities of this are given below:

• An NHS manager is likely to be responsible for a particular service across the whole NHS Board and can therefore find it difficult to develop partnership working with one Local Authority in a way that creates a 'postcode difference' in quality of care within the NHS board area. This was described as inhibiting more innovative work based on a Local Authority level, because it might be seen as creating postcode-differentiated care provision.
• A Local Authority service manager is responsible for services within one Local Authority area but has much more limited scope, to influence policy or practice, in neighbouring Local Authorities within the same NHS Board area. This was reported as making it more difficult for Local Authorities to engage with integrated work at an NHS Board level, without individual representation from each Local Authority involved.
• A GP works in independent practice with responsibility for a smaller population of patients. The independent nature of GP practices was reported to make it difficult for GPs to act in a representative way beyond the scope of their own practice. Engaging GPs in a NHS board-wide consideration of integrated working was, therefore, seen as requiring engagement with large numbers of GPs. This implied significant costs for locum cover in order to facilitate widespread attendance (especially if this was to be an ongoing dialogue rather than a one-off event).

Creating incentives to participate

5.27 The mapping information produced by the IRF and the new models of working piloted in the test sites are strengthening the evidence for changing the way in which services are planned and delivered. However, the IRF information and emerging examples were reported to be limited to effecting marginal change unless relevant stakeholders were given clear incentives to integrate and the power to influence more substantial change. Only 11 percent of service delivery stakeholders surveyed during the final round of the evaluation reported that the IRF had empowered frontline staff to improve services.²¹

5.28 The experience of the test sites highlighted four key factors which could incentivise professional staff to get involved: patient outcomes; professional standing; financial reward; and workload balance. Where these were present and recognised, engagement in service change was more likely. These are discussed in more detail below.

5.29 Patient outcomes - "Does this integration agenda improve the care of my patients/clients?" Stakeholders involved in the direct delivery of care reported that they prioritised changes where there was a direct link to the quality of care provided to the population for whom they had a caring responsibility. The evidence from the email survey of service delivery stakeholders was that over the period in which the IRF ran, their confidence that the IRF would directly affect patient/user outcomes dropped significantly from 6 to 3.²²

5.30 Professional standing - "Does this integration agenda make me better at my job or make my job more valued?" Many of those involved in service delivery (e.g. GPs and hospital clinicians) have chosen this as a vocation and have committed many years to training and skills development in order to practice their chosen profession. Where an integration agenda and process maintains clear professional boundaries, explicitly values the roles of different professions and provides security about future employment then service delivery staff may be incentivised to get involved. However, in the short term, the service change involved in integrating different services was reported as creating levels of uncertainty about new professional roles and responsibilities. For example, where integrated working seeks to shift the balance of care and potentially reduce the amount of care in acute settings this could be a disincentive for acute sector workers who may have concerns about fewer jobs and reduced responsibility.

5.31 Financial reward - "Does this integration agenda improve my financial standing (increased remuneration, where income is directly affected by activity or improved efficiency, where staff performance measures are partly based on ability to deliver within budget or achieve savings targets)?" In recent years, there has been a move towards using financial rewards to lever changes in the ways services are provided. For example, the General Medical Service (GMS) contract linked the Quality and Outcomes Framework (QOF) with the financial rewards available to General Practices. In recent years, efforts to improve the management of different parts of the Health and Social care systems have increased. For example, the GP contract created a greater degree of control over the working priorities of general practices by awarding QOF points (linked to financial payments) to a variety of behaviours seen as beneficial to patient outcomes. However, the experience of the test sites showed that initiatives such as the IRF, which are not directly included in such incentivised schemes, are unlikely to receive as much attention. It was reported that a national review of the GP contract may be the best way to do this. The IRF did not create any explicit links between engaging in the process of change and any sort of financial return. Stakeholders reported a lack of financial incentive to engage with the process of integration. Hence it was a lower priority than other activities which had an immediate and positive financial reward.

5.32 Workload balance - "Does this integration agenda make my work balance easier (reduce demands on my time or allow me to make better use of my time)?" One of the driving forces for integration is that services are currently under significant pressure to deliver the amount of care required by an aging population. Frontline staff are, therefore, already very busy and are looking for ways in which workloads can be decreased or better balanced. In this context it was difficult to find staff able to take time out of delivering care in order to look at longer term efficiencies and quality improvements.

5.33 One of the key issues highlighted by this analysis of the IRF test sites was the challenge posed by the short term barriers to effective participation. The experience of the test sites indicated that the early stages of trialling new integrated models of working may experience difficulties such as: lack of evidence to link proposed service changes and improvements to patient outcomes; increased professional uncertainty during a time of change; additional start-up costs to implement new ways of working while ensuring smooth service transition; and additional time commitment expectations placed on already busy professionals.

5.34 Results from the two rounds of the email survey indicated that, between 2010 and 2012, service delivery staff remained confident that their organisations were committed to the IRF.²³ However, these staff also reported decreasing levels of confidence that the IRF would improve joint working between health and social care.²⁴ Stakeholders reported that these short term disincentives should either be addressed directly or a much stronger link made with the anticipated benefits that could result in each of these areas in the medium or long term.

Key findings

• Test sites made quicker progress when they took a pragmatic approach to integration by building on existing local knowledge and relationships. These sites restricted their scope to a manageable geography, population, number of service partners and ambition of impact. An alternative approach taken in Highland indicated that more ambitious change could be introduced where local leadership was willing to extend the timeframe for implementing change.
• The IRF test sites found that it difficult to get meaningful representation or engagement with groups such as GPs or hospital clinicians. There are four key factors which could be used to incentivise greater engagement with delivery staff: patient outcomes; professional standing; financial reward; and workload balance. Where these were present and recognised then engagement in service change was more likely.
• The test sites did not implement financial mechanisms within the expected IRF timeframe due to a range of factors including announcements by the Scottish Government on next steps for policy on integration.
• The IRF raised partners' awareness variation in activity and in costs. However, it also brought into sharper focus the difficulties of improving efficiency and addressing unwanted variation when there is no direct financial mechanism to link demand and supply across the health and social care system.
• The IRF helped to support the case for shifting the balance of care by continually placing the idea of opportunity cost into discussions on the best use of resources. However, there was a lack of evidence that the IRF has influenced transfer of resources, and fixed costs remain a significant challenge.