HSJ Local Briefing is our in-depth analysis of key issues facing some of the health service. This week’s briefing looks at plans to integrate health and social care services in Somerset.




Issue: The ageing demographic make-up of the population of South Somerset is placing health and social care services under strain. The county council, district hospital, community provider and clinical commissioning group have set up the Symphony project to develop a model of integrated care intended to both improve services and boost efficiency.

Context: National policy is increasingly looking to integrate health and social care services to alleviate financial and demographic pressures. The Symphony project board has submitted a bid to become an integrated care pioneer under an initiative being led by health minster Norman Lamb.

Outcome: The project faces stiff competition from approximately 100 other areas of the country also keen to become pioneers. However, local commitment and enthusiasm for the project is strong, which should ensure it continues with or without pioneer status.

The key players and their aims

Discussions about moving to a more integrated model of care have been taking place between organisations involved in health and social care in the area covered by South Somerset District Council for the past year.

The five key players are Yeovil District Hospital Foundation Trust, Somerset Partnership Foundation Trust, Somerset County Council, Somerset Clinical Commissioning Group and the South West Commissioning Support Unit. They have each contributed £27,000 to fund the project’s development over the past 12 months. This has led to the creation of the Symphony project board.

The plan is to use an alliance contracting model, which will spread the risk and share any gains between all providers. However, this has never been tried before in the NHS.

Although the initial plans are to start small, focusing on a cohort of the sickest patients within South Somerset (population 164,000), the ambitions are to roll it out across the county. There is talk of eventually creating some form of accountable care organisation, which is paid using a per capita budget to incentivise improved quality at reduced cost. It is hoped the new model of care will be introduced for the initial group of patients by next April.

The driver of the proposals

Almost a quarter of the population of South Somerset is aged 65 or over, compared with 17 per cent in England as a whole. Between 2013 and 2021 the number of over-65s in the district is forecast to increase by 20 per cent, while the number of people aged 85 or over is predicted to rise by 30 per cent.

There is recognition locally of a need to do things differently to provide high quality care in the face of both demographic and financial pressure on services.

According to Yeovil District Hospital medical director Jon Howes, at any one time over 50 per cent of the hospital’s 345 beds are occupied by patients over 80. This population is set to double over the coming years.

“If everything stays the same we’d have to build another three wards to accommodate this demographic growth, yet we know at least 25 per cent of these patients can be managed more appropriately in another care setting,” he said.

The national context

Integration is becoming an increasing policy priority for the government, which views it as a means of offsetting the demographic and funding pressures facing health and social care services.

It is also hoped integration will improve the experience of the growing number of patients that have complex needs or a wide range of conditions − factors that bring them into regular contact with many different services.

Health minister Norman Lamb warned earlier this year the NHS was at risk of “collapse” if it did not do more to integrate health and social care. He therefore invited health economies to bid to become “integration pioneers”.  

The Department of Health document setting out the criteria pioneers must meet calls for the “most ambitious and visionary localities” to come forward with proposals. Successful projects will be provided with support and advice from a central integrated care and support exchange team, which is based at the Department of Health, for up to five years.

The document stated: “We will seek to address at local level any additional barriers that emerge as pioneers push forward and we will assess whether any rules should be changed at the national level, as a result.”

The DH received bids from 111 areas − a tremendous response considering there are only 152 top tier local authorities providing adult social care. Of these, 99 met the basic criteria; only about 10 will be selected in the first wave but it is possible two further waves of pioneer projects will be given approval.

South Somerset’s perception of integration

The project board favours the adoption of an alliance contracting model, in which all of the providers involved are represented on a board that holds a single contract with the commissioner. This way all organisations in the alliance are working to the same contract objectives and share the risks.

According to the Symphony project’s pioneer bid document, another plus is that it should provide the ability to move resources around the system, moving away from current situations in which “the additional costs sit in one organisation’s budgets and the benefits in the other”.

The model has been found to be successful in the oil industry and, although it has been attracting some interest in the UK health sector with discussion at a Nuffield Trust conference last year, it has not yet been tried in the NHS.

In a blog on the model, Paul Corrigan, who was health adviser to Tony Blair, details how it developed as a solution to costly building projects that overran and often resulted in the different providers suing each other. When alliance models were used, projects often came in ahead of time and under budget.

Robert McGough, partner in DAC Beachcroft’s health commercial team, described plans to introduce a new contracting model at the same time as a new model of care as “ambitious”. He told HSJ it was difficult to make direct comparisons with the traditional NHS contracting model as alliance contracts were outcome focused, rather than process focused.

He pointed to the difficulty in judging when an outcome had been achieved in a healthcare setting compared with a construction project. Mr McGough said the alliance model exchanged the “contractual legal certainty” of a standard NHS contract for a “looser form of arrangement” based on “good faith, integrity” and the shared interest in achieving the commissioned outcomes.

“This would be a significant shift in risk for a commissioner to accept and would be outwith of the guidance and structures which they are currently required to use,” he added.

Jeremy Martin, project director at Symphony project, told HSJ the project board had selected the alliance model − rather than other arrangements that involve one organisation taking the lead − due to a desire to work together. He said: “We’ve got a good history of working together in Somerset. There’s been quite a lot of stability despite recent organisational changes and quite a lot of relationships that have built up over time that are still in place.”

Selecting a cohort of patients who could benefit most from integrated care

The project board commissioned the South West Commissioning Support Unit and the Centre for Health Economics at the University of York to develop a database linking anonymised patient level data for more than 109,000 patients across primary, community, mental health and social care.

This included information on age, number of GP contacts, cost of acute care at national tariff, average cost per primary care encounter, community hospital bed days and the costs of any social care. Data on other key services such as district nursing and end of life care is not yet available but hopefully will be next year.

An analysis found the number of conditions an individual had was more likely to lead to higher costs of care than age. In total 11.45 per cent of the variation between money spent on individuals’ care was explained by how many conditions individuals had. By comparison, just 3.35 per cent of variation in costs could be attributed to age.

Mr Martin said the large amount of unaccounted for variation, which could be down to factors such as home circumstances, made it important to get the size of the first group of patients right. “The shared budget will have to have enough patients in it to cope with the variation between patients, but we also need to start with a small enough group that we can realistically start working in a new way over a short time period,” he said.

The annual cost per patient for an individual with seven or more conditions, £10,741, is 13 times that of a patient with just one disorder. The analysis found the majority of the difference is driven by acute inpatient costs. More details of this analysis can be found in the bid document.

Kevin Hudson, director of business solutions and innovation at the CSU, told HSJ the agreement of GPs to share practice level data had helped produce a “much richer view of the patient”. He said the data suggested a need to take a more “holistic” look at patient care, rather than focusing on disease specific approaches.

“This is going to be invaluable when other CCGs come to look at collaborative care and year of care budgets,” he added.

Andrew Street, professor of health economics at York University, said the dataset was “up there with the best” due to the linking of activity and cost data across health and social care organisations and the ability to look at the combination of conditions patients have.

The university is to host some workshops with health and social care staff in South Somerset in September to agree which groups of patients should be targeted.

Professor Street said: “Getting people together to talk about things and share experiences and understand the nature and pattern of care in the area will provide ownership and ideas of how these models might be rearranged for the benefit of the local population.”

Based on the data available, it was found that 2 per cent of the population accounted for £17m of the £100m costs across the health community. Dr Howes said: “These patients could be taken out of the current model of managing them through primary care and managed through some sort of network that integrates all the teams at community hospitals and possibly Yeovil District Hospital.”

A new financial model

The ambition is to develop a “year of care” style weighted per capita budget for the patients in the first cohort. Decisions are still to be made about how this would work in practice and how the money would move between organisations. Mr Martin acknowledged it will be “complicated” to negotiate the budget to the satisfaction of all parties.

Mr McGough, of DAC Beachcroft, said the normal method for an alliance contract would see participants paid for certain agreed costs incurred in delivering the service. A risk and reward regime would then incentivise improved performance against the outcomes in the contract.

He added: “Clearly, in order to implement an alliance contract model, then the finance and remuneration mechanism needs to be considered in detail to ensure there are no perverse results (excessive rewards or penalties imposed upon the parties) and that it is possible to utilise the alternative financial structure under the terms of NHS financing such as payment by results.”

Likelihood of success

There is a clear commitment from all partners to press ahead with the project, whether it gains pioneer status or not. However, the additional DH support that pioneer status would bring would undoubtedly prove valuable. Mr Martin sees the biggest advantage as the “support and expertise” to help the project through some of the contracting and financial issues.

At first glance the project meets key criteria set out by the DH. This states they must have a clear vision, involve planning for whole system integration, have support from across the community and commit to sharing lessons across the system.

Support across Somerset appears to be strong with organisations including: local housing association, Yarlington Housing Group; Yeovil College; and local charities South Somerset Mind and Age UK Somerset backing the project.

More than 60 staff from various levels of the different partners have been involved in discussions about how to integrate services while a separate event brought together 30 local GPs and consultants to discuss the plans.

Somerset local medical committee chair Sue Roberts told HSJ there was a lot of enthusiasm for the project. “Most of the Yeovil GPs are involved in the project and they all buy into it. It’s a new and exciting way forward,” she said.

Areas are also required to show they have the capability and expertise to deliver public sector change at “scale and at pace”. Two members of the project board were members of the senior leadership team at Torbay Care Trust, which has led the way in the development of integrated community health and social care in England.

Yeovil District Hospital Trust chief executive Paul Mears was director of operations at the trust for four years while Bristol, North Somerset, Somerset and South Gloucestershire local area team director Anthony Farnsworth spent six years at director level at the trust, including two and a half years as chief executive.

Somerset Partnership has form on the issue after integrating mental health and community services over the past two years. The organisation has also recently worked with the county council and private sector social care providers to set up independent living teams focusing on the reablement of patients.

In terms of the requirement to demonstrate the approach is based on a “robust understanding of the data”, the York University analysis has reportedly been described by academics there as one of the “richest” datasets of its kind.

The project also has influential support in the form of Yeovil MP and schools minister David Laws. However, the large number of bids means only one in every 10 has a chance of success and the competition includes many more established integration projects, most notably Torbay. With odds like that the Symphony project is far from a shoo-in.

The challenge of implementing an alliance contract should not be underestimated, with or without pioneer status. The rhetoric from the government surrounding the integration pioneers is radical − pioneers will be expected to have tested new reimbursement models. However, there is scepticism about how much freedom pioneers will actually be allowed.

The criteria document suggests there will not be scope for ignoring current choice and competition legislation but merely for “clarification of rules and how integrated solutions can comply with them”.

The level of enthusiasm among the various partners, particularly GPs, bodes well for the continuation of the project regardless. However, most of the toughest conversations about how the money and risk is shared between providers are still to be had.

The combination of the national policy focus on integration and strong local commitment to improving patient experience should ensure progress on the Symphony project but the partners may need to be flexible about what form that takes.