‘Integrated care’ has long been identified as one way of addressing the challenges associated with the increasing fragmentation and specialisation of care.1 Following Ara Darzi’s NHS Next Stage Review, the Department of Health launched its Integrated Care Pilots (ICPs) programme in April 2009 amid much hullaballoo.2 From a national invitation across England, 16 sites were chosen for participation. A major evaluation was expected to add significantly to the evidence base. Its final report was published in March this year3 but those seeking to find out whether integrated care ‘works’, let alone what works and when, are likely to be disappointed. This editorial examines findings from the evaluation.
AN ACADEMIC CHALLENGE
The ICPs, established under varying financial circumstances to meet different needs, must have presented a health services researcher’s nightmare. The most obvious problem besetting all investigation in this area is the lack of common definitions for concepts underlying integrated care.4 The plethora of different terms used is testimony in itself. The report’s authors provide useful frameworks for categorising integrated care but the reader may be left wondering whether this is not a term that has outlived its usefulness (Box 1).
Box 1. Types of integration5
Functional integration: coordination of key activities such as financial management, human resources, strategic planning, information management, and quality improvement.
Organisational integration: the creation of networks, mergers, contracting, or strategic alliances between healthcare institutions.
Professional integration: joint working, contracting, or strategic alliances between healthcare professionals within and between institutions and organisations.
Clinical integration: extent to which patient care services are coordinated across various personnel, functions activities, and operating units of the system.
The evaluation combined an intriguing mix of methods. Formative approaches included written feedback, regional events, teleconferences, feedback on the quantitative data collected, one-to-one discussions, and sharing evidence of good practice. The quantitative components included analyses of hospital usage data, surveys of users and staff at two time points, and an elaborate costings exercise. Financial data were particularly problematic and the authors acknowledged important limitations implicit in a ‘before and after’ study of this nature. However, the volume of activity and weight of evidence cannot conceal a disappointing lack of clear conclusions.
WHAT DID THE PILOTS ACHIEVE?
The ICPs implemented ‘a loose collection of integrating activities based on local circumstances’.3 All sites attempted the integration of practitioners working in different organisations. Most concentrated on the horizontal integration of community-based services such as general practice, community nursing, and social services. Most pilots were based in primary care and involved multiple partner organisations. They adopted an approach that identified specific populations, most commonly people at risk of emergency hospital admission. Interventions varied but a common feature was the use of a multidisciplinary team. The virtual ward, where patients were discussed but not present, was established on five sites with a case manager reporting back to other clinicians.
As to whether ICPs improved quality of care, the authors conclude that:
‘… if well-led and managed and tailored to local circumstances and patient needs, they can, but improvements are not likely to be evident in the short term’.3
These findings are more underwhelming given the considerable support ICPs enjoyed from both their status as Department of Health pilots and from the local evaluation team.
There was evidence of improved team working and communication within and between organisations. Changes to work patterns provided more interesting jobs.
However, patients did not generally share the enthusiasm of staff. They experienced more difficulty seeing the doctor or nurse of their choice following an intervention and reported being listened to less frequently and being less involved in their care. The authors attributed this to the professional, rather than user-driven, nature of changes and to unfulfilled expectations given the ambitious changes the pilot leaders set themselves.
Of course, the great hope of politicians was that ICPs would yield cost-efficiencies and, from that perspective, the most significant findings in this report are negative. There were no overall changes in the costs of secondary care use. There was a 2% increase in emergency admissions for pilot patients (though the unexpected increase may have been due to imperfect matches and controls) with a reduction in elective admissions and outpatients of 4% and 20% respectively. Only at the six case management sites focusing on patients at high risk of admission was there a net reduction in combined in-patient and outpatient costs. Even in much-feted Torbay, south-west England, reductions did not occur in the targeted older age group.
LESSONS FOR POLICY MAKERS
The complexity of integrated care activities should not be underestimated and can overwhelm even strong leadership and competent project management. Activities need to be matched to local capacity and change may take longer than anticipated. New services required up-front investments which were never likely to be recouped within the period of the pilot. The needs of users can easily be overlooked when building an organisational platform for integration. Successful integration of care is less about adhering to a particular model of delivery than finding multiple, creative ways of reorganising the work.
The facilitators and barriers to success identified might have dropped out of any change management primer. Facilitators include strong leadership, shared values and supportive professional attitudes. Barriers include changes to staff roles, interventional complexity, and turbulence resulting from NHS reorganisation. Legislative barriers to pooling budgets remain one of the biggest obstacles.
SIGNIFICANCE FOR GPs
These diverse projects yielded few generalisable findings but the report contains plenty to interest commissioners. The authors identified steps the pilots went through in trying better to integrate care (Box 2).
Box 2. Integrating activities3
Building governance and performance management systems:
for example, setting standards, establishing protocols, and lines of accountability
Developing the local business case for integrated care
for example, showing how integrated care would improve care, modelling, monitoring frameworks
Changing attitudes and behaviours
for example, engaging staff and service users, encouraging more responsibility by staff
Developing the necessary infrastructure
for example, IT, multidisciplinary team meetings
Establishing financial and support systems
for example, realigning incentives, establishing joint budgets, and accounting arrangements
Shared information technology was often a vital ingredient and the simple co-location of professionals from different organisations plainly assisted joint working. However, sharing patient records is not just a technological problem. Agreed approaches to note-taking, the language, and abbreviations used are also needed. Many staff had gained knowledge and skills simply through working with other professionals.
However, the costings suggest that setting up new projects should be less of a priority than trying to improve the status quo. (Where I work, for example, a plethora of new community-based services have been established with seemingly no more than well intentioned — usually specialist — enthusiasm to justify them.) Community services may be an area for clinical commissioning groups (CCGs) to make cost savings. CCGs should be wary of exaggerated claims to be increasing efficiency, whatever their face validity.6
STRENGTHS, WEAKNESSES, OPPORTUNITIES — AND THREATS
An accompanying literature review suggested three conclusions. First, there is no single solution to integrating care. Second, success depends on the context in which any initiative is introduced. Third, interventions designed to integrate care may improve its processes but not users’ experiences; they rarely reduce costs.7
While it is uncertain whether these pilots increased emergency admissions, it is unlikely that they reduced them. The increase in admissions may, in part, have been due to the identification of more patients at risk and needing admission to hospital. In accordance with the first law of planning: the supply of new services tends to uncover previously unmet needs and generate new demands. An association between case management and increased admissions has been seen before,8 though focusing on so-called ‘frequent flyers’ may be the most cost-efficient course.
The overwhelming sense from these pilots is of much tinkering around the edges. The ICPs needed far greater focus allowing tighter evaluation ever to have yielded much lasting learning. Proper integration at the boundaries of primary and social care requires new integrated, budgeting models.9 Similarly, contracts rewarding hospitals per episode clearly conflict with community-based models of care. Without a different range of financial incentives, cost-savings at these interfaces are likely to prove elusive.
The National Commissioning Board and Monitor may seek changes to regulations governing payment across care pathways, staff employment, and competition in furtherance of integration.10 The vision is of GP commissioners, able to provide as well as commission services, taking on the risk of capitated budgets for their populations and working in clinical partnerships alongside specialists and community health services.11
Finally for GPs, the rationale for ICPs is a chastening reminder that we are no longer so effective in our traditional role of coordinating patient care. We are in danger of becoming just one more entry point into someone else’s care pathway. If — or rather when — clinical commissioning groups struggle to deliver savings, the clamour for vertical integration of primary and secondary care providers along transatlantic lines will doubtless intensify once again.
Notes
Provenance
Commissioned; not externally peer-reviewed.
Competing interests
The author has declared no competing interests.
- © British Journal of General Practice 2012