With the multiple challenges of tightening NHS finances, an ageing population and the growing burden of chronic illnesses, integrated care is seen by some as a key tool for improving the efficiency and effectiveness of care.
Many people (including people at Nuffield Trust) argue that integration can lead to both better outcomes and experiences for those who use services. In 2008 the Department of Health sought applications from sites that were interested in developing new models of clinically-led integrated care.
16 sites were eventually selected as pilots for the integrated care programme. They varied widely in the types of integration that they aimed to develop, the populations and conditions they were focussed on, and their outcomes of interest.
The evaluation of these sites led by RAND Europe, Ernst & Young and the University of Cambridge used a range of methods but included some quantitative analyses of hospital activity undertaken by the Trust (one of our specialities at the moment).
Our analysis of changes in hospital use and the associated costs, used methods that we previously applied in the evaluation of the Partnership for Older People Projects (POPPs), to find matched controls from across England for each person who was involved in one of the integrated care pilots.
We then compared patterns of hospital use six months before and after receiving the integrated care intervention. We analysed data on hospital use for 11 sites, comparing over 8,500 people who received an integrated care intervention and over 40,000 matched controls.
Across all sites we found significant relative reductions in the number of elective admissions (a relative reduction of 0.04 admissions per person, equating to 347 fewer admissions over all participants in the six months after the intervention), and the number of outpatient attendances (0.2 attendances per person, equating to over 1,700 fewer over the six month period).
However, we found no evidence of a reduction in emergency admissions.
These activity differences were also reflected in secondary care costs. We found a non-significant, overall relative reduction in hospital costs of £57 per person after the start of the intervention, compared to the six months before – a four per cent fall.
For sites who implemented a case management approach we saw a significant nine per cent relative reduction in hospital costs of £223 per person.
The other elements of the evaluation included some interesting findings – notably the surveys of patients and staff showed somewhat contradictory results.
While staff were generally positive about the pilots, reporting greater job satisfaction and variety in their roles, patients felt that they were less involved in decisions about their care.
Like much of the literature around integrated care, the results of the evaluation of the pilots are complex (perhaps even puzzling and a bit disappointing).
Whilst we did find evidence of reduced planned hospital activity, particularly in sites attempting some form of case management, these were not easy to interpret.
It may be that this reflects better coordination of care or that care was moved from hospitals into the community. However, we did not see the hoped for reduction in emergency hospital activity.
Our evaluation only focussed on changes in hospital activity in the six months after people began receiving the integrated care intervention. It is possible that in the early stages of these schemes most of the focus was on setting up the pilots, and therefore on internal factors, rather than on the care given to patients.
It might be that changes in activity will be seen over a much longer timeframe as integrated services become more established.
As NHS funding becomes increasingly tight, there is a constant search for interventions that will reduce the demand for expensive unplanned hospital admissions.
There is clearly a great deal of interest in the potential of integrated care – and other approaches have emerged since these pilots were selected.
No doubt there are some who hoped that the national evaluation would provide a definitive answer – especially one that was supportive of integrated care and that promised a solution to the pressure of rising emergency care.
Given the variety of different schemes that was always going to be optimistic.
Perhaps the most important lesson is that we need to be realistic about the ambitions of these schemes, recognise that they have a range of objectives and realise these will take time – but we also need to continue monitoring whether our ambitions are being met.
This article has also been published on GPonline's Inside Commissioning blog.
Chitnis X (2012) ‘Integrated care – the picture remains unclear’. Nuffield Trust comment, 26 April 2012. https://www.nuffieldtrust.org.uk/news-item/integrated-care-the-picture-remains-unclear