Ernst and Young’s Richard Lewis and Rand Europe’s Martin Roland and Tom Ling evaluate 16 integrated care pilots begun in 2009, and what the evidence can tell us about integrating care in mainstream services today.

The NHS Next Stage Review of 2008 highlighted the need for more integrated care within the NHS and between the NHS and social care. Poor integration was identified as a barrier to effective patient care and one which led to poor patient experiences and excess costs to the care system.

As a consequence, 16 integrated care pilots were established in 2009 to test different ways of delivering integrated care and were subject to a formal evaluation which we discuss here. The lessons from these pilots remain relevant to today’s debates as “integrated care” has been highlighted as central to improving NHS services.

The 16 projects that formed the national integrated care pilot programme were highly varied in nature (see box, below). This variation was in part an intended outcome of the piloting process, with open applications for pilot status and with few constraints placed on the type of integration to be tested. 

The differences between the individual pilots did, of course, present challenges for the evaluation; pilots were sometimes trying to achieve quite different things in different local circumstances.

However, despite the variations across the pilots, they broadly shared a number of similar aims: bringing care closer to the service user, creating teams that crossed organisational boundaries, providing greater continuity of care, providing more preventive care and avoiding unnecessary hospital care.

In particular, a subset of six pilots focused on intensive case management of elderly people at risk of emergency hospital admission. They did this through the use of risk profiling and other tools to identify older people at risk of admission followed by the use of a case manager, most often a nurse, to coordinate the care of individual patients.

Most of the pilots concentrated on integrating community based services (such as general practice, community nursing and social services) and far fewer on ‘vertical’ integration between primary and secondary care.

The evaluation was designed to understand what whether integrated care offers benefits for patients, staff and the wider NHS. It drew on a number of different evaluation methods: “before and after” staff and patient questionnaires, analysis of hospital activity data and costs compared to matched controls, interviews with patients and staff, a regular journal compiled by the pilots themselves and non participant observation.

Staff across all pilot sites reported improvements in many important areas. Regarding overall patient care, 54 per cent of staff thought it had improved over the previous year, compared to 1 per cent who thought it had got worse.

Significant improvements in team working and communication were also reported; for those staff directly involved with the pilot, 60 per cent thought that they worked more closely with other team members, 51 per cent that communication had improved within their organisation and 72 per cent that communication had improved with other organisations.

These figures suggest that many of those involved in delivering integrated care had real confidence that it was providing benefit for service users.

Staff also reported changes to the nature of their own work as a result of the pilot. For example, 84 per cent of those directly working with the pilots felt their job had expanded and 63 per cent thought that it had become more interesting.

Staff also felt that integrated care increased efficiency and saved money, including a reduction in emergency admissions. However, analysis of hospital utilisation data showed a somewhat different picture to that described by pilot participants. Emergency admissions for pilot patients were actually higher than for the control group. Some of this increase in admissions may well have been due to imperfect matching of the control group (a known hazard of this type of evaluation), however, even taking this into account it is very unlikely that emergency admissions were reduced by the pilots.

Outpatient attendances and elective admissions, however, did reduce as a consequence of the pilots. This was particularly the case in the subset of pilots that implemented case management, where statistically significant falls were detected (22 per cent and 21 per cent for outpatients and elective admissions respectively).

Applying cost estimates to these utilisation changes suggests that for all pilot sites there was a non-significant reduction in overall secondary care costs. However, for the case management sites a significant reduction in overall secondary care costs of 9 per cent was identified. In both cases, this does not take into any increased costs in primary or community care which may offset these savings.

Perhaps the most surprising results concerned patient views of integrated care as practised by the pilots. While staff felt confident that their care had improved, patients reported a more mixed picture. More technical aspects of care were perceived to have improved, with significant increases of 8 per cent in patients receiving a care plan and 9 per cent in those knowing who to contact with questions after hospital discharge.

However, these positive results were balanced by a number of negative findings. For example, there were significant reductions in patients being involved by their doctors in decisions about their care and of seeing the nurse of their choice of 5 per cent and 9 per cent respectively. Moreover, there was a significant drop of 15 per cent in those patients feeling that their opinions and preferences were taken into account by their care workers. These negative findings were generally more evident in the case management sites.

The issue of poor care integration has remained close to the top of the policy agenda over the three years since the pilots were established. The Future Forum, established to advise on the Government’s reforms of the NHS, reiterated this concern, highlighting poor integration as a key faultline in the NHS and wider care system. So these evaluation results provide grounds for optimism that integrated care will deliver at least some of the hoped for benefits.

The experience of staff within the pilots has been positive – with a clear belief that integration has improved the way in which they deliver care. There is also evidence that particular types of integrated care, specifically those focused on “case management”, can deliver overall reductions in the costs of secondary care.

However, this optimism must only be cautious at this stage. Change takes time and, as more than a third of staff surveyed suggested, it may simply be too early to tell about the full impact of integrated care on outcomes. More surprising is the reaction of patients to integrated care. While care processes such as care planning may have increased, important aspects of the patient experience have diminished. One hypothesis is that integrated care within the pilots became “professionalised” and, in the process, lost focus on the individual patients at the centre of that care.

There is a further concern for policy makers if integrated care is to be promulgated at scale – the 16 pilots were enthusiastic volunteers. Moreover, they were volunteers who were supported by a national programme, project management support and funding. Even these enthusiasts struggled at times to cope with the demands of leading change in a difficult and shifting environment. Translating the efforts of a few into the actions of many will be the challenge if integrated care is to shift from pilot to mainstream.

Summary of objectives by pilot site:

PilotMain integration focus / client group
Bournemouth and PooleStructured care for dementia
Cambridge AssuraEnd of life care
Church View, SunderlandElderly people at risk of admission
North CornwallMental health care
CumbriaPeople at risk of admission (self-management)
Durham Dalesa) Rapid access medical assessment clinic with reclassification of acute hospital as community hospital, b) Moving services closer to home, c) Fuel poverty intervention, d) Improved transport to services, e) Older people’s mental health
Nene (Northamptonshire Integrated Care Partnership)People at risk of admission to hospital (long-term conditions)
NewquayStructured care for dementia
NorfolkLong-term conditions
North TynesideFalls in over-60s
NorthumbriaChronic obstructive pulmonary disease (COPD)
Principia, Nottinghamshirea) People at risk of admission b) COPD
Tameside & Glossopa) People at risk of cardiovascular disease (CVD) b) People with CVD
Torbaya) Prevention of admission of elderly to hospital b) Enhanced discharge planning, c) People in nursing homes with COPD/ congestive cardiac failure (CCF) d) Services for low-level dementia
Tower HamletsStructured care for diabetes
WakefieldSubstance misuse