Long-term care for older people in Jersey - Professor Julien Forder

Long-term care for older

people in Jersey

Prepared for the Health, Social Security and Housing Scrutiny Panel, States of Jersey

Julien Forder November 2008

Julien Forder PSSRU

University of Kent

Contents

1. Introduction ....................................................................................................................................................................... 3 The future of long-term care........................................................................................................................................... 3
2. Long-term care provision............................................................................................................................................ 5 Range of services .................................................................................................................................................................. 5 Developing long-term care .............................................................................................................................................. 7 How does Jersey fare? ........................................................................................................................................................ 8 Developing services in Jersey ...................................................................................................................................... 11
3. Delivering the required care................................................................................................................................... 14 How does Jersey fare? ..................................................................................................................................................... 18
4. Funding and funding arrangements ................................................................................................................... 21 Demand in the future ....................................................................................................................................................... 21 Current funding arrangements................................................................................................................................... 22 Changing the way we pay for care ............................................................................................................................ 23 Options for Jersey .............................................................................................................................................................. 27
5. Summary and conclusion ......................................................................................................................................... 32 Appendix 1. Design issues ............................................................................................................................................. 37
6. References ........................................................................................................................................................................ 39

1. Introduction

Jersey, like many jurisdictions and countries, faces some difficult choices as regards long-term care. Demographic pressures combined with people's growing aspirations about quality and choice will increasingly pressure the current system. Many countries have embarked on more fundamental reform of long-term care, concerning both the funding and delivery of care.

Long-term care is one of the remaining great uninsured risks that people face. For those unfortunate enough to need formal care services – and the lifetime risks of needing some care at 65 could be as large as 50% for women – the cost consequences are stark. If people pay out of pocket – and many do – the cost can easily run to over £1000 per week if residential care is required.

Long-term care supports people to deal with the consequences of physical or cognitive impairments, rather than trying to address the causes. Often there are no effective health care interventions that can tackle any underlying health problem – for example, people with dementia – beyond some mitigation of symptoms. But people also need help with the disabling effects of their conditions; help with activities of daily living, such as feeding, toileting, dressing, washing; help with practical tasks, such as shopping and cleaning, and to an increasing extent, support to improve quality of life, reduce social exclusion etc. (Knapp 1984; Norton 2000; Brodsky, Habib et al. 2003). This is the role of long-term care. And it is a significant role. Across the OECD, public expenditure on long-term care homes and home care services averaged over 1% of total economic output of each country in 2000 (OECD 2005) .

This paper reflects on the situation of long-term care in Jersey and draws on evidence gathered from a series of (public) hearings with ministers, key officials, independent stakeholders, user organisations and provider organisations. The Panel also visited Guernsey to learn about the long-term care insurance system that had been in operation there since 2003.

The future of long-term care

Academic, policy and other work in the UK and internationally highlights some important trends that will influence the nature of long-term care in the future (OECD 2005).

• Needs and aspirations. People's life expectancy is increasing and looks set to continue increasing well into the future. Older people will form a greater proportion of the population in the years to come. Developments in medicine have been successful in keeping people alive for longer. What is less clear is whether healthy life expectancy is increasing at the same rate. In any case, without a significant compression of morbidity', we expect to see many more older people with care needs in the future.  

At the same time family composition is changing. There is increased female participation in the workforce. People are more socially and geographically mobile. Predictions are hard to make, but the suggestion is that rates of informal caring by family members will fall in the future.

People's attitudes to care are also changing so that service users in the future will be less accepting of poorer quality care. People want greater independence and more choice. People will also want support that helps them to live with a good quality of life, rather than just survive. In other words, as well as physiological well-being and health, psychological and social well-being are important to people.

• Technology and skills. Care services are highly labour intensive. But the increasing cost of labour and falling price of technology will likely drive a reduced labour intensity in the future. Monitoring and warning technologies –telecare' – can potentially reduce the role of care staff in the supervision of confused people. New skills and approaches are developing that corresponding to new ways of caring for people, for example, in caring for people with dementia.
• Financial context. Both individual finances and the broader fiscal situation of jurisdictions and countries are changing. Older people in the future will be wealthier than the current cohort, particularly as a result of the significant growth in the value of housing assets in the decade before 2005. There is also (arguably) a resistance in the population to paying higher rates of taxation. Public authorities are becoming more aware of the opportunity costs of their actions, in the context of limited resources. There is greater emphasis on ensuring that the system offers best value in delivering it aims and objectives.  

These three sets of factors are likely to imply a different package of services and support should be available in the long-term care system compared to what has been provided in the past. In order for the long-term care system to deliver these changes, the right processes and funding needs to be in place. Again, there are discernable trends in this regard:

• Decision makers are becoming more keyed in to what people want from their services and support. There is a greater emphasis on the outcomes people want i.e. social participation, purpose, control and safety as well as physical and environmental comfort; what care needs people have; and how these outcomes and needs can be measured (Forder, Netten et al. 2007). Decision makers are investing in the knowledge and evidence required to give a good understanding of what forms of care are needed.

• There are innovations in how the care system is organised and managed; for example, having well-defined referral routes; comprehensive assessment of need; robust independent regulation; and so on (Knapp, Hardy et al. 2001; Forder 2002).  
• Many countries are re-considering how to pay for care. They are questioning the balance of funding between state and individual, between means-tested and universal funding; and between funding against risk (in advance) and funding at the point of need (Poole 2006).

Public long-term care systems are complex and multi-faceted. Achieving the appropriate package of support and services for people involves: knowing what the right package looks like; having the right systems in place; and, ensuring the required funding is available. Against this backdrop the paper assesses the direction of long-term care policy in Jersey in three areas, contrasting the Jersey experience with innovations outlined above.

• First, it considers the range and balance of services and support available for older people in Jersey.
• Second , it explores how the long-term care system is organised and managed.
• Third, the paper describes and assesses the funding of long-term care in Jersey.

2. Long-term care provision Range of services

Long-term care for older people comprises a range of services that can be distinguished according to the needs of service users. It is also useful to distinguish care with housing from community-based care that occurs in people's own homes. The range of options is as follows:

Care with housing

• Continuing care, which is nursing home care for very high dependency patients, often with severe dementia and challenging behaviour, whose needs straddle the (blurred) boundary between social care and on-going health/nursing care. This form of care is often provided by the public health care system rather than social (welfare/security) system.

• Nursing home care, covering high dependency social care need. Nursing homes cater for people with relatively stable underlying health conditions but who need very intensive social care support and also nursing aimed at mitigating as far as possible the impairing consequences of, or progression of, chronic disease.
• Residential care, which is primarily for social care needs dealing with the disabling consequences of impairment. Residential care serves people that can no longer remain in their own homes because of risk or a poorly suited housing environment.
• Residential care with enhanced care, which is residential care but which caters for people with more challenging consequences of their condition e.g. advanced dementia that can result in aggressive behaviour or poor communication skills.
• Extra-care housing, which is specialised housing in a complex where an on-site care team provides domiciliary care. People in these schemes usually live in flats with their own front-doors in a complex which has communal facilities such as a restaurant.
• Sheltered housing, which refers to private housing (sometimes purpose built) in a complex where there can be warden support but not an on-site care team.

Community-based care

• Home nursing and home care. Social care and community nursing provided to people in their own homes. A distinction is often made between personal' care and practical' or domestic care. The former concerns services that help people with activities of daily living (ADLs) such as washing, dressing, feeding and so on. Practical care is about maintaining a person's environment, but where there is no linked health risk.
• Respite care, which covers temporary stays in a care home to provide a break for family carers.
• Day care, which is where people live in their own homes but attend day centres for meals, social support and sometimes personal care.

In addition, there are a range of health services (often hospital based) that provide clinical and professional support to older people in long-term care e.g. consultant geriatricians in outreach teams. Professional therapy services, such as physiotherapy and occupational therapy (OT) tend to straddle the boundary between health and social care. There are also hospital out-reach teams catering for people with mental health problems (including Old Age Psychiatry).

Intermediate care is a form of short-term care (usually up to 6 or 8 weeks) with a focus on rehabilitation of those people discharged from (acute) hospital care either back home or into long-term care (Department of Health 2001; Melis, Olde Rikkert et al. 2004). Another form is rapid response. These are services that try to reduce inappropriate admission to hospital using a short burst of intensive support, deployed quickly, and diverting patients from A&E. The evidence is not completely unequivocal but suggests there are net benefits from the use of intermediate care (Godfrey, Keen et al. 2005; Barton , Bryan et al. 2006).

Intermediate care tends to focus on physiological problems or health-related rehabilitation. Reablement' services, by contrast, seek to help people adjust back to living independent lives, particular after a stay in institutional care. The focus is more on regaining life skills and confidence rather than physical functioning (Pilkington 2008).

General housing for pensioners and older people is also of relevance to the (low end) of the long-term care spectrum. Lifetime homes are designed with access and features that support people to live at home for longer, when and if a care need develops.

Developing long-term care

The trends outlined in the introduction combine with the evidence-base to suggest the following aims for support and services in the long-term care system:

• First, services need to support people in their own homes for longer. In Scandinavian countries like Denmark and Sweden (and also in the UK), there has been significant development of intensive home care services that provide personal care to people in their own homes, with care packages ranging from 10 hours per week to 24-hour live- in care (Wanless, Forder et al. 2006). This model sees a reduced role for traditional residential care.
• Second, with the increase in longevity in the population, there is an increasing need for services to support people with very high levels/complex needs, especially as concerned with severe cognitive impairment. Important aspects of dementia care include (Godfrey, Surr et al. 2005; Wanless, Forder et al. 2006):

• continuity in care staff, so that the person with dementia is not unsettled by regular changes in domiciliary care staff
• staff with specific training in dementia care
• an emphasis on maintaining physical health, despite the mental deterioration
• high-quality day care centres for leisure and social contact
• memory clinics' – effectively a one-stop-shop' offering assessment, diagnosis, support and counselling, information, monitoring of treatment, and education and training

• regular respite care as part of a package of measures to relieve the burden on informal carers.

• Third, technological developments offer the possibility of a greater use of telecare and other forms of electronic assistive technology (Poole 2006).
• Fourth, there should be an increased use of rehabilitation, reablement and intermediate care services (Glendinning and Newbronner 2008).
• Fifth, a greater role for extra-care housing, particularly as a substitute for lower- intensity residential care (Netten, Darton et al. 2008). The opportunities for extra care to supplant higher dependency nursing home care are more limited however.

How does Jersey fare?

The following account gives an overview of the main service options available in Jersey. The aim is not to give a detailed, exhaustive account of all facilities but to provide a summary of the main types of care for older people with long-term care needs. This summary sets the scene for an assessment and discussion of current provision and potential new developments.

Residential and nursing home care

Table 1 (below) gives a breakdown of residential and nursing home provision in Jersey. The table lists registered places in private, voluntary and Parish homes for older people and also public sector nursing places. These are places registered for older people. The table also distinguishes continuing care in Jersey in public sector nursing homes. Continuing care services are also out-sourced to private sector nursing homes.

The number of residents at any time will vary, but we can assume that occupancy rates generally exceed 90% for independent & parish sector homes and approaching 100% for public nursing places. In this case, the total number of residents in 2007 was around 850. This is equivalent to around 65 residents per 1000 population 65 years or more, or 140 people per 1000 population over 75.

Community-based care

The vast majority of home-based social care that is funded by the States is out-sourced to Family Nursing and Home Care (FNHC), a very well established charitable organisation. FNHC provide a range of services including: district nursing; home care; professional therapy services (such as occupational therapy); and children and family services. Home care is an increasingly important form of long-term care for older people. In 2007, FNHC provided home care services to around 2200 people.

Table 1. Provision of residential and nursing care - number of registered places 2007 for older people

Type of care  Number of places Independent & parish sector  

Residential care  Parish  216 Voluntary  168

Private  292

Total Beds older people  676 Nursing Homes  Parish  30 Voluntary  5

Private  188

Total  223

Total Beds older people  203

All independent/Parish  Total Beds older people  879

Public sector owned continuing care/nursing beds

Generic  Limes  36 Sandybrook  28

Total  64 For people with mental  Oak  26 health problems  Maple  25 Lavender  10

Total  61

Care is divided into three categories: practical and domestic care (level 1); personal care (level 2) and twilight care. In 2007 the break-down was as follows.

• Level 1: 951 people
• Level 2: 1063 people
• Twilight: 224 people

Some people received more than one of these types of home care. Level 2 and Twilight services

are focused on people with higher levels of need. Taken together, FNHC provided Level 2 and

Twilight services to around 100 people per 1000 population over 65. An age breakdown is not available, but the majority of these people will be over 65. As such, the rate of use in the population is slightly higher than the rate of use of care home services (see above).

A distinction is often made between intensive and regular home care, where the former is seen as a realistic alternative to a care home placement. It normally entails at least one visit per day to the service user. In 2007 total visits for all forms of home care ran to nearly 110,000, which implies an average of about 1 visit per week. Although the number of patients with complex needs have increased, FNHC indicated that their service is mostly focused on moderate (personal care) needs. In 2008 some 13% of clients received care 7 days per week and another 8% received care 5 days per week; the remaining 79% received care less frequently (ruling out intensive personal care that is required daily)[1].

A number of private agencies also provide home care and home nursing services, as do the community mental health service and individual private carers. These services clearly add to the total level of community-based care provision, but FNHC is the main provider. In addition to home care, FNHC provided district nursing services to just under 1000 people over 65 in 2007. The physiotherapy and OT teams visited 2120 people.

Around 75% of FNHC funding is from a (non-specific) grant from the States (£5.8m). Membership fees and charges for home care and medical supplies add another £1m, with a further £1m coming from donations, fund-raising and other sources. Under this funding arrangement, referrals are made directly to FNHC who undertake an assessment of the client and determine a care plan. Around 50% of referrals are from GPs, 40% from acute hospital and 10% from other sources (such as direct user referral). There are plans to move from the grant to a service level agreement and this is considered in the next section.

FNHC report that service delivery at current levels is only achievable because the organisation is also supported by charitable and donated funds. In particular, current funding precludes expansion into intermediate care, such as rapid response and hospital at home.

FNHC also see a priority in increasing the flexibility and choice available within the home care service. At present, service users have limited choice over the time slots for home care visits, meaning that people may be put to bed earlier than they would like. There are currently no provisions in Jersey for people to take cash alternatives (so called, direct payments or personal budgets) rather than services in kind – Section 3 below has more details.

The main form of respite care that is available for older people is short-stay placements in care homes. There are currently 7 respite beds available. The geriatric mental health service also has a small number of respite places.

Regarding day care, HSSD has three purpose built day centres and can arrange day care at other centres such as those run by Good Companions.

Developing services in Jersey

At present in Jersey there appears to be a relatively high level of traditional' residential care, with a relative dearth of alternatives like home care, and sheltered and extra-care housing. In England, in 2006/7, just under 340,000 older people were in either residential care, nursing care or residential continuing care (two-fifths of people are state supported in care homes). This number corresponds to around 40 residents per 1000 population over 65 years and 85 residents per 1000 population over 75 years old. Making explicit comparisons are not without definitional problems, but the rates in Jersey appear to be about half as much again as in England. When account is made of the lower average wealth of people in England and so higher rates of disability, the discrepancy with Jersey would be even greater.

Over the last 10 years, there has been a marked increase in the numbers of hours of home care provided in England. Between 1998 and 2007 the number of hours of home care funded by the public sector (via councils) increased by just under 50%. At the same time the number of households supported fell as the home care hours were targeted on people with higher levels of dependency. In 2007, 52 per cent of households received more than 5 hours of home care and 6 or more visits per week and 30% of households received more than 10 hours and 6 or more visits per week (The Information Centre 2008). Between 1998 and 2007 the latter number of households increased by 70%. The number of individuals over 65 who received state supported home care in England during the year 2006/7 was 480,000. Something like 100,000 more people purchased home care privately (Forder 2007). This translates to about 70 persons per 1000 population over 65 years of age

In 2007 FNHC in Jersey provided personal (level 2) home care to around 80 persons per 1000 population over 65 (and to more people if level 1 care is also included). This rate is slightly higher than the England rate, but appears to be aimed at lower dependency people on average. Data from FNHC shows that the number of households receiving intensive support (i.e. daily visits) is significantly less than the England case. FNHC provide daily (7 days per week) visits to 13% of clients; in England in 2007, 45% of people received 10 or more visits per week and a further 24% received 6 to 9 visits per week. The extent to which home care is used as a substitute for residential care in Jersey looks to be more limited than in England.

A recommendation therefore is for the provision of intensive home care services to be increased. Resources would ultimately come from reducing the number of people in residential care. In the meantime, some double running costs would be inevitable. In England, a target of increasing the ratio of people supported in intensive home care as a proportion of the total number of people in residential care and intensive home care by 1 percentage point was consistently achieved by local authorities since the early 2000s.

A further recommendation is for the development of intermediate care services on the Island. FNHC felt that a pilot rapid response scheme was successful. The evidence elsewhere is not unequivocal but strong enough to underpin a recommendation for a States-funded pilot programme.

The view of most respondents in the hearings was that the level of sheltered housing in Jersey is limited. At present there are no extra-care facilities available. Following the above arguments, development of extra-care would seem to offer a good fit in terms of catering for the apparent gap in provision as between lower dependency residential care, and domiciliary care for people remaining in their own homes.

Extra-care is not a cheap option (Baumker, Netten et al. 2008), not least because the floor space of flats in extra-care schemes is greater than for rooms in care homes. Furthermore, funding is needed to meet the costs of the communal facilities that are available (e.g. leisure facilities, lounges, cafes, laundries) and also the warden, maintenance and other support functions. In England, housing benefit is available to meet the basic rent, and council social services funding is used to meet the care costs. The costs of support and communal facilities can be met by the user, from a specific public funding initiative from central Government called supporting people', or from other housing related benefits. Supporting people has played an important role in a number of publicly-funded extra-care schemes, filling the funding gap between the housing and care components of the overall package (King 2001; Darton, Baumker et al. 2008).

The extra care housing schemes in England offer a variety of housing tenures, allowing people to safeguard their financial assets by purchasing or part-purchasing their accommodation. Tenure models include market or social rent, buy and shared ownership units and in many cases, a mix of these arrangements. Early indications from the research are that "in terms of social climate, the mix of residents with different tenures is something that may need to be carefully managed" (Darton, Baumker et al. 2008, p. 78). For example, the social mix can

become segregated along tenure lines. There are questions about whether wealthier people would buy into schemes with a high proportion of social rents. Nonetheless, having shared ownership options were seen as important by the majority of residents in the study, allowing people to downsize and release some equity but also maintain a capital stake in the property.

Where people remain in their own homes, low level equipment and also more advanced assistive technology could have a role to play. A ring-fenced grant may serve to pump-prime uptake of new technologies. Basic equipment includes grab rails, raised toilet seats, accessible baths and so on. More advanced equipment includes telecare devices such as falls detectors, blood pressure monitors, smart home technologies that monitor entry and exit, gas or bath monitors, automatic lighting etc. (Audit Commission 2004). Early evidence suggests that these technologies have potential to reduce avoidable hospitalisation, to improve home security, to allow virtual visiting that in turn improves both well-being and good health. Social alarm systems can also provide feedback and instil confidence to help older people retain their independence. The cost-effectiveness of these technologies is less well established, but generally the unit cost of these technologies is low, and so the potential for assistive technology to substitute at least in part for expensive labour inputs appears high. Jersey already has systems in place (e.g. alarm technologies) and would be well placed to implement the more advanced technologies, especially as the unit costs of this technology fall.

Given the likely demands on care services in the future – see Section 4 below – and the on-going problem of finding sufficient workforce capacity to a required standard and cost, policies to mitigate unnecessary use of high-labour-intensive care options need to be considered. Re- ablement and rehabilitation services can potentially help to delay the need for high-intensity care. In England, an evaluation of a series of pilot re-ablement services is being undertaken[2]. Re- ablement uses short-term bursts of support by care workers and occupational and physio- therapists to help people regain skills and confidence as regards basic activities of daily living like washing, dressing, eating, mobility etc. Preliminary findings from the research are that re- ablement services can reduce the need for future home care.

Jersey retains a significant level of continuing nursing home care provision in public sector homes (i.e. the Limes and Sandybrook). There has been a program of closing old public provision (e.g. McKinstry) and the out-sourcing of this high dependency provision to the private sector. The primary motivation appears to have been the quality and standards of the public provision. There is some indication that The Limes is also showing signs of age in this respect (e.g. inappropriate client-group mix and also physical build issues such as narrow corridors, limited sight-lines and multi-storey design). Given the aging of the population and projections about future rates of disability, we would expect demand for high-dependency care services to grow. There may be capacity in the private sector (although it is expensive), and this needs to be determined, but otherwise, further investment will be required to enhance or newly-build public provision. In any case, having a mix of public and private provision is likely to balance concerns about the security of provision with pressures to keep costs low.

Although a whole range of cost and other implications would need to be investigated, development of the Overdale site might be considered. This site could potentially accommodate extra care housing units and also a high dependency nursing home. These close care' arrangements help to provide continuity of care and allow people easier transition between more and less intensive care options as their conditions change. In many cases the stereotype of gradual decline during old age does not apply, with instead people under-going short term crises from which a degree of recovery and improvement is quite possible.

It is also recommended that dementia care services on the Island are further developed in line with expected future increases in need. In most cases, people with advanced cognitive impairment will require residential forms of care, rather than home care, to provide the significant levels of supervision that are needed.

Although these recommendations would change the balance of services in Jersey, freeing up some resources, overall they are not cost neutral. The higher cost would, nonetheless, be accompanied by a significant improvement in care provided in Jersey. Section 4, below, considers how additional costs could be met, but ultimately there would need to be a willingness of the general population in Jersey to pay more for this improved care.

3. Delivering the required care

There are a number of key processes undertaken by the long-term care system in delivering services and support to people with care needs. They determine how:

• people are referred;
• their needs are assessed;
• their care plans are determined;
• the required services and support are commissioned;

• services are provided;
• and how quality and standards are regulated.

The way these processes are undertaken varies considerably between different countries. Nonetheless, a number of important policy choices can be discerned. The first is whether long- term care is largely organised as a health responsibility or a social welfare responsibility. In many countries a distinction is made between health care and long-term care and the issue is how well integrated these systems are in practice. For example, in the Scandinavian countries there is a very high degree of integration between health and social services. Other countries show far less integration. In England there is a strong separation: local councils organise social care and the NHS runs health care.

Second, whether there is a separation between the agency that assesses a person's care needs and the agency that is responsible for making available public funding. There are broadly two alternatives (Forder 2002; Brodsky, Habib et al. 2003). The entitlement model has a strong separation – clinicians make judgements of need and third-party payers such as insurance funds or social security agencies provide the funding (subject to financial eligibility rules). The German system is an example. The budget-constrained model, in contrast, has the same agency responsible for care planning and funding. The English care management system is an example. Care managers located in councils decide service needs but can explicitly take into account the available budget when making their decisions.

Third, whether the focus of needs assessment is clinical or social. The former approach tends to concentrate on the nature of the person's impairment, whilst the latter focuses on the ability of the person to lead their life as usual. For example, the former might highlight lack of mobility, whilst the latter might assess the person's resultant limitations in terms of their social life and social participation (Netten and Davies 1990). Mostly a clinical assessment is required, but additional information on social needs might help to improve the allocation of services (Forder, Netten et al. 2007).

A fourth distinction rests on whether systems offer cash payments in lieu of services. In the Austrian system, for example, long-term care benefits are paid as cash leaving recipients to determine how to use these resources. In Japan, which also has a long-term care insurance system, care managers decide services and cash alternatives are not available.

A fifth distinction concerns the way people access the care system. Some arrangements allow self-referral; some allow general practitioners to determine access to care services on people's behalf, and others require all people to be assessed by a gatekeeper'.

What is current thinking? As people live longer and potentially have longer and more complex care journeys, there is a need to ensure a high level of coordination between care options. Although the evidence is patchy, there are a number of policy implications (Wanless, Forder et al. 2006, esp. chapter 4):

• Comprehensive (single) assessment. Assessment should cover social, health and housing needs together with robust assessment tools to avoid duplication and inconsistency. Self-assessment can also play an important part. Assessment should also be made in terms of people's potential to benefit from services – in a comprehensive sense – as well as in terms of their current (clinical) needs.
• Coordinated care planning. Decisions about support and services need to be made with sufficient information and where a full range of service options is available. Service users should be at the heart of the process. Inevitably decisions about placements will be influenced by the current service capacity, but the service users' needs and preferences should predominate over supply considerations.
• Commissioning. Public authorities have a key role in commissioning and securing the service capacity to meet expected need. There should be the flexibility to shift funding between service options in response to the prevailing distribution of need. In practice, where different organisations or departments with their own lines of accountability and funding are responsible for different service options (as in Jersey), this flexibility and co- ordination is undermined. A silo' mentality can inappropriately limit service availability. In systems where money follows the service user, these problems are minimised. A separation of commissioner and provider can also give service users more choice and help ensure that they are not subject to provider capture' (Forder, Robinson et al. 2005).
• User choice and personalisation. It is easy for public systems to take insufficient account of the preferences of service users, and this can result in poor outcomes for people. There are many ways that people's preferences and choice can be built-in, but the research literature indicates that service users are empowered when they have control over budgets. There are many examples internationally of long-term care systems that give people cash payments or individual budgets in lieu of services. There is as yet limited, but growing evidence that this improves their outcomes, particularly as regards to feeling in control of one's life (Glendinning, Challis et al. 2008).

The following (highly summarised) account describes the relevant processes in Jersey. Referrals to the long-term care system in Jersey come from the hospital, GPs and also self-referral to Social Services or directly to FNHC and private provider agencies. Where people present to the public system an assessment is undertaken. This can occur at a number of different points in the system, but a placement tool is available which acts as a template to standardise assessment outcomes. The placement tool rates the severity of people's needs in a range of categories including, nutrition, incontinence, medication, personal care, mobility, memory, depression/anxiety, challenging behaviour, pain, sensory impairment and a number of health conditions. A scoring algorithm and in some instances a case conference then determines a service need as one of the following:

• Supported home care;
• Residential care;
• High dependency residential care;
• High dependency mental health residential care;
• Nursing care;
• High dependency nursing care for older people with mental health problems;
• High dependency continuing nursing care;

Like most assessment tools used to date, the placement tool used in Jersey is mainly clinically orientated. Future development could also include assessment of the person's potential to achieve well-being outcomes such as social participation/loneliness, occupation, living environment (Forder, Netten et al. 2007). Assessment tools should also account for the potential preventative effect of services. In other words, where people's current needs are relatively low, but there is suggestion that matters could worsen (e.g. a break-down of informal care arrangements), this should be addressed.

According to the design of the Jersey tool, case managers are assigned to undertake this placement assessment and then manage the referral-to-final-placement process. Departmental responsibility depends on which of the above service options is required. Most home care is managed by FNHC.

People needing residential care can apply to the Social Security Department (SSD) for means- tested public funding support, or refer themselves privately. In the former case, provisional eligibility for SSD support in residential care is predicated on the result of the placement tool analysis. If the tool indicates a residential care need, then the amount of financial help from social security is determined by the financial means-test (as outlined below). In theory, people can choose from the range of private, Parish and voluntary sector homes in Jersey. Fee rates

differ between homes and are met by service users, drawing on SSD support where relevant. In practice, where people are reliant on SSD support, they are limited in their choice according to the fee that SSD is prepared to pay. At present, the SSD has not made explicit its tariff of prices. Some provider organisations stated in the public hearings that the fee that the SSD were willing to pay did not meet the full costs of care. They also highlighted that some providers were receiving higher payments than others.

As with residential care, people can either privately refer to nursing home care in an independent nursing home or approach the Health and Social Services Department (HSSD), where they are assessed as above. People qualifying for standard, EMI or continuing nursing care are then placed by HSSD in a public home (e.g. the Limes), a contracted-out bed or a spot- purchased bed in a private nursing home. Placements in any of these settings are funded by the HSSD although a means-tested accommodation charge is made to the resident. Some people will pay the full charge (just over £420 per week) and some will receive public funding to help pay this charge. At present (Aug, 2008) there are 47 contracted beds i.e. where HSSD purchases the place for a period of time, and 30 spot purchased beds where the bed is purchased only while the named person remains a resident of the home.

Through choice or to avoid waiting lists, some people directly approach nursing homes rather than HSSD. In this case, they are liable for the whole fee, rather than just paying an accommodation charge.

HSSD manage (acute) hospital services as well as long-term nursing care placements. This contrasts with other systems (e.g. in England) where long-term care is primarily a social services responsibility and hospital care is a health (NHS) responsibility. In England, the implication of this shift of responsibility has been that service users now receive means-tested support in the social care system rather than the more generous funding support in the health system.

At present private and voluntary care homes are regulated by the Health Protection Department. Reforms are in train to extend inspection to cover non-residential care providers, including FNHC. The legislative basis for inspection is currently the 1984 Act. Providers receive two unannounced inspection visits per year. Inspection reports are shared with providers but are not publicly available.

How does Jersey fare?

There is a separation of responsibility and accountability between home care, residential care and nursing home care. The placement tool helps to make patient's needs explicit in the system, but resources do not automatically follow service users. Also, service users face different charge regimes between these three service options. Furthermore, as regards residential care, the HSSD is in a position to influence demand (via application of the placement tool and other preventative health measures) but low-income people are entitled to funding support via the social security system. In turn, the Social Security Department, in trying to manage expenditure, has only the direct lever of the prices it negotiates with providers – it cannot directly manage client-side demand. There is also the residual role of the Parish welfare system to further complicate the picture.

Such a situation can lead to some people receiving inappropriate care, to cost shunting and perverse incentives. It will also lead to significant price pressure on providers. Both these consequences are evident in Jersey – an apparent over-use of standard residential care and also strong downwards pressure on prices.

In England in 1993 reforms brought care home and home care funding streams together as a local authority responsibility. Residential care funding moved from social security to local authority control (Wistow, Knapp et al. 1994; Wistow, Knapp et al. 1996). Subsequently there was a significant re-balancing of residential versus home care as described in Section 2. In many cases, service planning decisions were pushed down to front-line care management staff. Although, care managers still complain about being financial gatekeepers', they are in a good position to make informed trade-offs between individual service user needs and the cost implications of service decisions. Individual or personal budget arrangements – if rolled out across England – will shift the budget even closer to the service user.

It is recommended that a greater alignment of funding streams be sought in Jersey. One option would be to set up a single commissioning body in Jersey that brings together the funding and procurement of States funded nursing homes, residential and home care. This commissioning function would operate at arms-length from providers, both public and private, and should improve resource flows and ultimately outcomes for service users. Such a body would need to form strategic partnerships with the acute health sector in order to mitigate any potential problems of bed-blocking.

A more radical option would be to shift social security funds to HSSD. This would align budgets but would have the disadvantage of requiring HSSD to means-test individuals for entitlement to public support.

A third option would be to build on the placement tool to develop it into a fully blown commissioning tool. In this way, people's needs and potential outcomes could be assessed and a

point scoring system used to capture the results e.g. more points means greater need. The

second step would be to determine the monetary value of these points for the procuring of services and support. Ideally, the authorities would calculate the expected total points across the expected caseload in the next period and then divide the available global budget by this points total to give a price per point'. The price would then be multiplied by the number of points each user was assessed to have. From this budget per individual, any mix of services could be commissioned from any provider type. A further recommendation is that this individual budget could be made available as a real cash payment if the service users wished.

Whatever the exact form of the commissioning function, benefits are likely to result from drawing clear lines of demarcation between commissioning and providing. Services could then be procured using service level agreements (perhaps on a cost-and-volume basis) in order to share cost risks. The recommendations made in Section 2 imply an increased role for FNHC. It would seem appropriate to ensure the accountability of FNHC through a tighter contractual arrangement than is currently the case. Indications are from FNHC that this would be welcomed; the financial risks associated with varying referral rates would then be shared between FNHC and HSSD.[3]

As regards the fees that Social Security are prepared to meet, or indeed any commissioning body that operates in its stead, an argument can be made for a more transparent payment structure. An explicit tariff of prices rather than individual provider fee negotiation may be seen as fairer and would reduce the costs of commissioning. On the other hand, provider costs do differ on a provider-by-provider basis for legitimate reasons, and pricing should sometimes accommodate these differences. Also, a fixed tariff of fees based on client type will encourage up-coding' where clients are assessed up to the next tariff point. Whatever the choice, because this concerns public funding, agreed prices should be a matter of public information.

Better co-ordination with housing is also recommended. One option is to set up joint projects around the development of extra-care housing.

The proposed extension of regulation to all providers – residential and community-based, private and public – is welcomed. The regulator would ensure standards of care for all providers. It is recommended, nonetheless, that lighter touch' forms of regulation be explored, such as inspection holidays for consistently high performing providers. This might entail primary legislation. In addition, the focus of inspection (in legislation) should be shifted more towards service user outcomes. The 2000 National Minimum Standards (NMS) in England provide some guidance, but a better template from England is the Key Lines of Regulatory Activity (KLORA) used by the Commission for Social Care Inspection (CSCI) in England.[4]

The Social Security Department are aiming to make improvements in these areas. In their draft amendment, Social Security are calling for a more complete' solution, involving:

• Standard care contracts to protect all individuals in residential care
• A range of standard fee rates for residential homes accepting States supported residents
• More effort to improve provision of care in the community

4. Funding and funding arrangements Demand in the future

Along with many developed economies, the demand for long-term care in Jersey looks set to increase considerably in the next 20 years. Life expectancy is increasing rapidly, which will mean a greater number of older people in the population in the future. Recent population projections for Jersey (Oxera, 2007) suggest that the number of people over 65 will increase by around 50% over the next 20 years, a figure very much in line with similar projections for England. Table 2 gives the population of Jersey and projected numbers in 20 years time.

Table 2. Population estimates

Age group  2001  2007  2027 Over 65  12,330  12,841  19200 Over 75  5,692  5,928  -

Source: States of Jersey Statistics Unit and Oxera (2007)

Jersey is relatively wealthy and therefore we might expect healthy life expectancy to increase at a rate faster than England. Nonetheless, in most scenarios age- and gender-specific rates of disability do not fall by much, if at all. Improvements in cardio-vascular health will likely be overshadowed by higher rates of diabetes and obesity, for example. The incidence of dementia is closely associated with age, not how close people are to death. Furthermore, the unit cost of services is increasing. Most estimates point to unit costs continuing to increase at 2% ahead of price inflation in the future, a rate consistent with earnings growth.

Based on the modelling done for England, the total cost of providing an equivalent level of support to individuals in 20 years time could double in real terms. And this is assuming that the amount of support people get at any time remains the same as now. In section 2, by contrast, the argument was made that support packages should be improved. In some cases, they may be cheaper, but overall it is not difficult to imagine that the costs of care packages will increase as quality rises and outcomes improve.

There is no particular reason to expect costs to increase at a much lower rate in Jersey. It will therefore be more difficult to fund long-term care as a residual claim on public revenues without significantly increasing taxation or social security rates. Public resistance to increases in general taxation rates will then present two alternatives as future costs rise. Either a system of specific contributions is introduced or people pay a greater share out-of-pocket.

Current funding arrangements

At present the funding arrangement for long-term care in Jersey are complicated and multifaceted. There is a different system for (public) nursing care homes, residential care, home care and other community services. The following is a brief overview.

People needing (continuing) nursing home care have the care element of their service covered by HSSD funding, but are expected to pay a charge for accommodation and subsidence. This arrangement covers people over 65 in public beds (e.g. the Limes) and also people placed in contracted-out and spot-purchased beds (e.g. at Silver Springs). The charge is just over £420 per week (2008). People unable to afford the charge can apply for financial support, which is subject to a means-test. Currently a little over a half of people receive public funding support, with the remainder paying the full accommodation charge. People who approach nursing homes privately (without going through the public system) pay the full home fee as negotiated privately.

People assessed as needing residential care are charged the relevant care home fee (which is set by the home). Those people unable to afford this rate can claim income support. People receive income support equal to the shortfall of their assessable income (above a minimum personal allowance) relative to the home fee. Where people have eligible assets above a threshold, these assets must be used to pay the full cost of care before income support funding can be received. In other words, income support is only paid out after a person has run assets down below the relevant threshold. Assets may only become available in the estate of the resident after they die. In this case, income support payments that have been made in the interim may need to be repaid. In both the residential and nursing home case, income support is only payable if people are placed according to assessment and subsequent application of the placement tool.

People receiving home care pay an annual membership subscription to FNHC (of around £50 per year). They may also face modest charges at the point of use.

The income support system in Jersey is currently being reformed. At the time of writing (2008), Social Security plan to reform the treatment of income support for people in residential care. One option under consideration is to develop a long-term care insurance scheme, although no details are available as yet.

Changing the way we pay for care

Design issues

Contemporary thinking about reforming funding systems for long-term care concerns a number of key design questions, including:

• What the balance should be between collective contributions (individuals paying into a fund in advance of needing care) and people paying out-of-pocket for their care support at the point of need';
• whether the scheme is re-distributive or risk-adjusted. The latter means that people pay contributions set according to their lifetime expected cost of care. The former involves cross-subsidisation between people;
• whether the scheme is mandatory or voluntary;
• whether the scheme is pay-as-you-go (PAYG) or funded. In the former, contributions cover the costs of people currently needing care. A funded system has people make contributions in order to cover their own risk of needing care in the future (using some form of saving mechanism);
• how benefits (services or support) to individuals are determined relative to their needs;

• whether or not the scheme is carer blind' i.e. if the assessment for formal support or services accounts for informal care or not;
• the degree of choice people are given e.g. whether cash payments are available instead of services.

Further details and assessment of these design issues can be found in Appendix 1 below. A developing model

The above design questions indicate the wide range of possibilities for configuring a long-term care funding system. Drawing on international experience, we can identify a small number of models, as follows.

• Family support. In this model most long-term care is provided by family, friends or the local community. For people without these carers the charitable sector might be available (e.g. the church), funding formal care from donated income. Otherwise, people pay out of pocket for formal care services. Although family care is hugely important in all countries, few developed nations completely lack some form of collective solution. For poorer countries, this if often the prevailing model.
• Means-testing. In this model public funded support (from general taxation) is available to people who cannot afford to pay privately for care, on a means-tested basis. Wealthier people who do not meet the means-test are obliged to pay privately or rely on family support. The means-test is usually stringent, giving people just enough support for them to meet the costs of care. It is budget-constrained, with waiting lists or needs-based eligibility thresholds used to manage demand. Most of these models have a history in welfare or poor law equivalents, with a strong link to the social security or social assistance system. Countries like England, US, and Jersey use this model.
• Universal public support. Long-term care, like health care, is funded (and largely provided) by the public sector, essentially free at the point of use for all people regardless of their financial means, with contributions coming from (usually progressive) general taxation. Scandinavian countries like Denmark and Sweden exemplify this model. Assessment of need is made by public authorities, often based on local professional judgement and is budget-limited (waiting lists are used when services are over-subscribed). Scotland has recently adopted this model.
• Social insurance partnership. Rather than general tax funding, long-term care can be financed by an earmarked contribution or premium, often a hypothecated payroll tax on

earnings. This usually goes hand-in-hand with making access to care an entitlement, rather than budget-limited. In theory, social insurance can cover the whole cost of care, but in practice this rarely happens. Instead a co-payment is required from the service user, although this co-payment charge is normally means-tested. Where this co-payment is relatively large, these arrangements can be described as a (point-of-need) partnership between individual and State. The long-term care systems of Germany, Japan, and Austria fall largely into this category.

Of these models, there is growing consensus that the first two arrangements are not sustainable, nor desirable in light of the trends outlined in the introduction. With aging populations, where older people are wealthier and more politically conscious than before, access to good quality care that does not threaten to impoverish people is now a much higher priority than in the past. In particular, most countries that are now categorised as in the latter two categories above have only relatively recently reformed their system to this model, having been in either of the first two categories before. Germany and Japan moved from a means-tested system, as did Guernsey. Spain is now adopting a social insurance model, whereas before had a family model with a small means-tested safety net. Scotland moved from means-testing to the universal public support model in the form of their free personal care scheme. In England, the Government has embarked on process to develop a Green Paper (expected in Spring 2009), which is principally about reforming the prevailing means-tested system. The rationale for those changes away from a means-tested model appears to hold strongly for the Jersey case. In particular, the means- testing model has the following problems.

• It is complex and could create perverse incentives or cost-shunting.
• People with high incomes and/or (eligible) assets will generally pay the full costs of care. With little availability of private insurance people are unable to insure against loss of property assets should they need care.
• As with many means-tested systems, funding of care may be seen as unfair – people who save are penalised. There is also the problem of incentives to fraudulently divest of assets.
• Because some people face the full costs of care (and have to draw down on assets), they may delay seeking care as long as possible, living with unmet' need.
• Being a safety net system, support is only available via social security for people who are impoverished'. This poor law' philosophy undermines dignity.

What system should replace means-testing? In light of the arguments made above about design, the scheme should have:

• a universal collective' component – all people get some help regardless of their means;
• a small co-payment that would be means tested;
• the universal collective component either funded from general taxation, or if the political resistance is too great, then an earmarked social insurance premium – and in any case, have only limited risk adjustment to allow re-distribution to favour low income, high risk groups;
• support for private insurance for people cover the costs of topping-up;
• cover for all forms of care, regardless of the care setting (i.e. not distinguishing care in a care home from care in a person's own home);
• regard to consumer choice, perhaps allowing people to take support as a cash payment;
• explicit eligibility criteria, with an assessment-based gatekeeper function.

Before considering these principles in the Jersey case, we can look at the highly relevant experience in Guernsey.

The Guernsey scheme

The Guernsey scheme is an example of a social insurance model. Given the clear parallels between the economic, social and geographical situations in Guernsey and in Jersey, the Guernsey experience of LTC insurance can give some useful steers as to such a development in Jersey. The main features of the Guernsey scheme are:

• a social security based insurance mechanism where all adults pay a fixed proportion (currently 1.4% on the first £60,000) of income or earnings (for the employed population).
• The scheme pays £341 for residential care and £637 for nursing plus a co-payment made by the recipient of £154 per week (this rate is also the charge for public nursing care not covered by the scheme).
• The co-payment is means-tested on income and savings, but not housing assets.
• Eligibility is determined by an assessment panel.
• Individuals contract directly with homes having secured a payout from the scheme.
• Home care is not covered in the scheme, although personal (home) care has no charge. Practical home help is chargeable subject to a means-test.
• There is no co-payment on respite care.
• The scheme replaced an apparently piecemeal and incoherent set of previous arrangements.

The system has been in operation since 2003 and is currently showing a yearly surplus. This is expected to diminish after full transition from the old system.

Guernsey's system has a number of advantages, but also, arguably, some shortcomings. The advantages:

• It provides insurance for all people;
• People are assessed for eligibility;
• The system is well-funding ;
• Contributions are made on income (protecting assets, which is the goal of the system);
• Improved funding has helped to reduce the problem of delayed discharge from hospital;

There are a number of disadvantages:

• The scheme only covers the cost of residential care – which can lead to under-utilisation of both low-level home care and high end nursing care;
• It only pays for private sector providers;
• The scheme acts as a third-party insurer – which offers few incentives for cost control;
• Any shortfall in payment against fees needs to be made up by top-up payments;
• With the contribution set as a capped proportion of income (and just earnings for employed people) the system appears to be regressive – poorer people end up paying a higher proportion of their total income to the scheme than richer people (but are more likely to use the scheme)
• The scheme is likely to get much more costly in the future

Options for Jersey

Accepting the arguments that the Jersey means-tested system needs reforming, what appear to be the most beneficial options? One option is to move towards a universal public model, like the Scott ish free personal care (FPC) scheme. In England the scope for this seems limited for three principle reasons, despite the fact that FPC scores well in most areas that a means-tested system is weak. First, given where we are now, providing personal care free of charge (or at least with much higher subsidies) arguably transfers too much public funding from poor to rich. Given that modestly rich people (but not the super-rich) do very badly under means-testing, the rationale for additional help for these middle income groups is compelling. But does FPC go too far in this regard? Second, without means-tested charges, there is prospect that people seek too much' support. Care managers will ration care according to need, but will be unsupported by the deterrent effect of small, means-tested charges. The third problem is the most significant. Free

personal care requires a great deal more public tax financing than the current system. For

example, in England, for people over 65, a move to free personal care would require a more than 40% increase in public funding. This increase alone would add 2.5% to the basic rate of National Insurance in England. The political feasibility of such an increase appears to be very low in the current economic climate.

This leaves a social insurance/partnership model – a hybrid solution. But what are the specifics? The learning from Guernsey is highly relevant and an option would be to implement a Guernsey style model having dealt with some of the shortcomings – this would be a Jersey long-term care insurance scheme. Even after dealing with the shortcomings, problems remain – an alternative is a partnership social insurance scheme. We consider these two options in turn.

1. A Jersey long-term care insurance scheme

A long-term care insurance scheme in Jersey would need to cover the risk of needing non- residential care. The Social Security Department has done some initial costing of this scheme based on just residential care coverage. Updating these, the gross costs in Jersey of residential and nursing home care is approximately £28m per year. If this total expenditure requirement was added to the social security contribution, as a pay-as-you-go (PAYG) arrangement, this would require the base contribution rate to increase by just under 2 percentage points (based on analysis by the UK Government Actuaries Department) and revenue base figures from a report by Oxera (2007).

These are gross costs – but a sizeable proportion of these costs are already covered by the public sector (perhaps approaching a half). Also, the total would include hotel costs' which might be made chargeable to the resident (subject to a means-test), in a similar fashion to the Guernsey scheme (where the resident pays the first £154 per week). The net additional required contribution would then be smaller in this case (i.e. savings could be made elsewhere that would partially offset the cost).

Adding non-residential care coverage would push up costs. At present the costs of community and home care services to HSSD runs to around £7.2m for people over 65. This amount may also turn out to be too small (see the recommendations made in Section 2), and certainly would be too small if the scheme needed to be carer blind (see above). Also, some part of the greater use of home care would be as a result of a lower use of residential care. Overall, nonetheless, both residential and home care for over 65s presently would cost in excess of £35m a year. Exact calculations would be required which factor in all these calculations, but the order of magnitude of required additional contribution would be at least 2 percentage points on the base social security contribution rate.

This estimate is for a pay-as-you-go arrangement. In this case, contributions for each year must cover the costs of care in that year. They will need to increase significantly going forward into the future (e.g. care costs doubling in real terms in the next 20+ years). At the same time, the revenue base for social security contributions could fall (Oxera 2007). Overall, this would mean that contribution rates would have to increase significantly over the next 20 years and beyond.

These amounts that people would have to contribute are sizeable whatever the exact assumptions, and will increase in the future. But they are high because the cost of care is high (and increasing). The problem is that the overall cost is greater than most people expect, which makes the contributions look expensive. Indeed, because an insurance scheme might have to be carer blind, it is possible that with limited co-payments (e.g. covering hotel cost only), it could end up costing more than a free personal care scheme (which is the case for equivalent calculations for England). It is worth noting that in the German case, people have the incentive to choose a (lower-value) cash benefit rather than services because cash can be used to pay informal carers. So although the system is carer blind, people with informal carers are more likely to choose lower value cash benefits, thus reducing the overall cost of the system. There is also the question of whether people trust governments to deliver on the implicit contract (i.e. when they are old, someone else will be asked to pay for their care).

Measures to make people more aware of the costs (e.g. independent analysis, awareness- raising) would help. If contributions can be specifically ring-fenced, people might be more prepared to pay, but the political resistance will still be considerable. The issue of trust (in Government) could, as for example in the German case, be addressed by making benefit and contributions decisions quasi-independent and/or matters for parliament directly.

Both the German and Japanese long-term insurance schemes have come under considerable cost pressure. The money value of payouts in the German system and also in the Scott ish system have been held constant since implementation (and so have fallen in real terms). The German system has also seen a substantial increase in the contribution rate since its inception a decade ago.

The Guernsey system has only been in operation for a short while, but is showing a healthy surplus (as was the German system at that stage, incidentally). It may also be more isolated from growing demand because it covers just residential care at a time when most extra demand growth is in the non-residential sector.

2. Partnership social insurance for Jersey

A partnership social insurance scheme reduces the amount that would need to be raised through the contributions system. Rather than seeking to cover the full costs of care in advance, the insurance system can be implemented to cover a part of the service or support cost, with the remainder paid at the time of use. For example, the insurance scheme could cover the first two- thirds of the cost, allowing people to top-up for the remaining third (or more) if they wished. This arrangement is sufficient to give people some assurance that they could secure some minimum level of care from the scheme (i.e. 67% of average value of the ideal care package) without having to pay anything further if that were their preference. The coverage rate would be means-tested so that people on lower incomes would have a higher proportion of care costs covered. It would therefore be a progressive universal' arrangement; all people would get some public insurance, but the rich would pay more (in higher point-of-need charges and also in contributions if they were non-regressive). As in France, people could also seek private insurance to cover these point-of-need charges if they wished. The French system has seen a very significant increase in people taking out these secondary private insurance policies.

An example illustrates how this system works. Take an 85 year old person assessed to need of £100 of care per week. If the person were wealthy the social insurance scheme would pay £67 per week, leaving the person to pay £33. If they were on low incomes, some part of the co- payment would also be covered by the social insurance system: instead of £67, the social insurance might pay £85, leaving the person to only pay £15, for instance. People would also be encouraged to take out private insurance to cover the risk of needing care and therefore facing a co-payment (the £33 or less). Suppose that the 85 year old person took out a policy at 65 with coverage of up to £100 a week (to pay the co-payment). In this example, this person would be completely covered against the costs of care, through a combination of social and private insurance.

The social insurance component would be mandatory and taken as an earmarked social security contribution. The private top-up insurance would be voluntary, but people would be encouraged to take it up. One option would be to invite all people at 65 to join an insurance plan, or even to auto-enrol them and have people actively opt-out if they so wished. Another idea would be to regulate the available private insurance to a small number of standard plans. Private insurance would need to be risk adjusted', that is, the contributions people pay would need to be broadly in line with their expected risk. Otherwise, low risk people would simply opt- out. Risk adjustment might pose some political problems – for example, women would pay much higher premiums than men as a result of have greater life expectancy. Hybrid solutions of this type are advantageous in many ways, but can be complicated for the general public – the private insurance component would have to be made as simple as possible.

There are further variants to this model, which could either introduce more choice or lower cost. For example, the social insurance coverage rate could be 50% of costs, not 67% as above. Or people could be invited to choose a coverage rate e.g. 50%, 67% and 90%, although this would significantly complicate how contributions would work.

Insurance schemes generally need to be carer-blind; they need to be based on entitlement criteria that is independently verifiable' (i.e. by a court) and this is difficult with regard to measuring informal care inputs. Nonetheless, markers are available, such as whether or not people live alone or with someone who could act as a carer; or whether they are married or not.[5]

The public sector cost of this sort of scheme will vary according to the way it is set up, but if it were only partially carer-blind and if the social insurance coverage rate was low, then the extra public sector cost would be much less than that of a full blown long-term insurance model.

Who should pay and how?

The appropriate choice of funding system for Jersey will in part depend on preferences regarding the protection of assets, and in particular, housing assets. Under the current means- testing system, people moving into residential care are often obliged to draw on housing assets. If the person remains in long-term care for a significant period of time then these housing assets can be completely eroded.

A move to a more universal financial system, where housing wealth is largely disregarded, will protect people against this risk. But any increased public funding for people that need care and at present would draw on housing assets, would have to be funded by increased contributions from other people. Tax-based and social insurance systems would generate extra contributions from (a) younger adults as well as older people, and (b) from non-service recipient older people as well as those with care needs.

In a situation where many working age adults are already under significant pressure in the amount of taxes and other contributions they are asked to make, the first option (a) is difficult. Instead, more could be sought from all older people. Although the system might want to protect people from losing all their assets in the worst case scenario, this is different from asking people to contribute a fixed sum or proportion of their assets in return for protection against

catastrophic risk. Indeed, there is significant value tied up in housing assets; and despite the recent economic downturn, most holders of housing assets have seen the value of those assets increase substantially in the past decade or more, a windfall that in many cases was tax free.

An option is to implement an insurance system only for older people, where premiums are secured from all people at (or over) 65 years of age. The premium could be deferred so that it is paid from people's estate. The main problem with these options is that the premiums appear to be very high. The average lifetime costs of care for all people at 65 have been calculated to be around £35,000 or more in England (this covers residential and non-residential care costs) and this amount would be the fair premium' of such an insurance arrangement.[6]

At present there would appear to be little appetite for people to pay these kinds of sums to cover care-related risks. For example, the private insurance market for these products is extremely modest. Limiting all contributions to older people rather than all tax payers seem unlikely to be a workable policy option on its own. Rather, a proportion of the cost would have to be met through general taxation or social insurance with a small old-age premium being paid at 65. The specifics could be debated, but such a scheme would draw on wealth tied up in housing assets, whilst not placing the whole cost burden on this source of funding.

Some argue that the problem with asset draw-down is that people are obliged to sell the family home when they perhaps need only a proportion of its value. A number of equity release schemes are available in the private finance sector for people to convert some of their housing wealth into income. These can be appropriate in certain situations, but tend to be expensive (Poole 2006).

5. Summary and conclusion

Jersey faces a number of difficult policy decisions concerning long-term care. The aging population, the costs of services, and the increasing demands from potential service users concerning quality and support in the future are heaping pressure on current arrangements. At the same time the technology of care is developing. Older people are wealthier than previous cohorts, but there is also increasing resistance in the general population to extra taxation.

This paper reflects on long-term care in Jersey. It draws on evidence gathered from a series of (public) hearings with ministers, key officials, independent stakeholders, user organisations and provider organisations, and also a number of visits to care homes and to Guernsey to learn about the long-term care insurance system that had been in operation there since 2003.

Policy with regard to long-term care has three challenges.

• First, achieving the right range and balance of services and support for older people;
• Second , deciding how the long-term care system is organised and managed to deliver these services;
• Third, putting in place a set of funding arrangements with which to pay for services and support.

Drawing on international experience and evidence, the paper outlines current thinking as to the best configuration of policy in these areas. These are compared with the current arrangements in Jersey and recommendations are made accordingly.

In summary, the main observations and recommendations about the Jersey situation are as follows.

Services and support

• A comparison of the use of residential, nursing home and home care for people over 65 in Jersey with England and elsewhere suggests an over-use of residential care. The numbers of people receiving home care is relatively high, but there appears to be an under-development of intensive forms of home care that could act as a real substitute for residential care.
• A main recommendation therefore is for the provision of intensive home care services to be increased. Resources would ultimately come from reducing the number of people in residential care. In the meantime, some double running costs would be inevitable. In England, a target of increasing the ratio of people supported in intensive home care as a proportion of the total number of people in residential care and intensive home care, by 1 percentage point, was consistently surpassed by local authorities since the early 2000s.
• A further recommendation is for the development of intermediate care (IC) services on the Island. The evidence of the performance of IC is good, but not unequivocal, so a gradual development (or comprehensive piloting) is suggested.
• The view of most respondents in the hearings was that the level of sheltered housing in Jersey is limited, especially very-sheltered or extra-care housing' facilities. States support for the development of extra-care schemes should be considered because these

services offer a good fit by catering for the apparent gap in provision between lower dependency residential care, and domiciliary care for people remaining in their own homes. Extra-care is not a cheap option, however, and although costs would be recouped from reduced use of residential care, extra costs would fall on individuals. A range of tenure options would help allowing people to retain a capital stake in the property.

• Where people remain in their own homes, equipment and also more advanced assistive technology will have a role to play. A ring-fenced grant may serve to pump-prime uptake of new technologies.
• Jersey retains a significant level of continuing nursing home provision in public sector homes (i.e. the Limes and Sandybrook) and contracted-out to the private sector. This is expensive care but is catering for people with the most intense long-term care needs. In the future this type of care is likely to be in greater demand and plans will be needed as and when the public provision becomes out-dated. Further contracting out seems to be part of the solution, but also retaining some public provision would be prudent in a small market place.
• It is also recommended that dementia care services on the Island are further developed in line with expected future increases in need. In most cases, people with advanced cognitive behaviour – that requires significant levels of supervision – will require residential forms of care, rather than home care.
• Although these recommendations would change the balance of services in Jersey, freeing up some resources, overall these proposals are not cost neutral. The higher cost would, nonetheless, be accompanied by a significant improvement in the care provided for older people in Jersey. New funding options could offer additional funding, but ultimately there needs to be a willingness of the general population in Jersey to pay more for this improved care.

Organisation

• The integration between hospital and long-term care in the Jersey system is a strong feature, but there is a separation of responsibility and accountability between home care, residential care and nursing home care. The placement tool helps to improve consistency, but resources do not automatically follow service users and people face different charge regimes for different service options. Furthermore, there is no explicit separation within HSSD of commissioning and providing functions.

• Two options might be consider. It is recommended that a single commissioning body is set up that brings together the funding and procurement of States funded nursing homes, residential and home care. This commissioning function should operate at arms- length from providers (in both the public and private sectors). It would use unified assessment and placement frameworks, including the existing placement tool. Such a body would need to form strategic partnerships with the acute health sector to ensure good co-ordination and continuity of care. The second option – that could either work with the current arrangements or with a new commissioning body – is to develop the placement tool into a full-blown commissioning system that links resources to care choices.
• Services would then be procured using service level agreements (perhaps on a cost-and- volume basis). In particular, in securing community-based services from FNHC, an SLA is required that better shares the risks between HSSD and FNHC. This is being planned.
• A further recommendation is that each individual, following assessment, is given an indicative budget for care and support, based on their needs. The States of Jersey should explore the possibility of allowing the person to take the indicative budget as a cash payment if they wished.
• The proposed extension of regulation to all providers – residential and community- based, private and public – is welcomed. It is recommended, furthermore, that lighter touch' forms of regulation be explored, such as giving inspection holidays for consistently high performing providers. In addition, the focus of inspection (in legislation) should be shifted more towards service user outcomes.

Funding

• Jersey operates a means-testing' model for the funding of long-term care, where public funding is available to support people who cannot otherwise pay for the care they need.
• There is growing consensus that this funding model is not sustainable, nor desirable in light of the trends outlined above. With aging populations, where older people are wealthier and also more politically conscious than before, access to good quality care that does not threaten to impoverish people, is a much higher priority than in the past.
• A number of countries have recently moved away from a means-testing model e.g. Germany, Japan, Guernsey, France, with Spain in the process. They have instead implemented a social insurance/partnership model where all people are (at least

partially) covered against the risk of costly long-term care, with funding coming from an

earmarked contribution made by the general population. Scotland has moved from means-testing to a universal public support model in the form of their free personal care scheme. In England, the Government has embarked on process to develop a Green Paper, which is principally about reforming the prevailing means-tested system. The rationale for moving away from a means-tested model expressed by these countries appears to hold strongly for the Jersey case.

• The States of Jersey in New Directions has announced a willingness to consider a long- term insurance system. The Guernsey model is a useful starting point, but would have to be extended to cover non-residential care. Even if it were set-up as a pay-as-you-go model, this is an expensive option. The new contributions would be high and their acceptability to the people of Jersey is in question, despite the attractiveness of a system that helps people protect their assets.
• The recommendation is therefore to consider an alternative – a social insurance partnership scheme. It reduces the amount that would need to be raised through the contributions system. Rather than seek to cover the full costs of care in advance, the insurance system can be implemented to cover a part of the service or support cost, with the remainder paid (out-of-pocket) at the time of use. For example, the insurance scheme could cover a minimum of the first two-thirds of the cost, allowing people to top-up for the remainder if they wished. The proportion covered by the scheme would be means-tested so that people on lower incomes would have a higher proportion of care costs covered. This arrangement is sufficient to give people some assurance that they could secure some minimum level of care from the scheme without having to pay anything further if they wished.
• Furthermore, the States would provide a framework for people to secure standardised private insurance to cover these top-up charges. The similar French system has seen a very significant increase in people taking out these secondary private insurance policies.
• The system would be financed through the social security contribution. In addition, policy makers might wish to consider the introduction of a small additional old age premium. This would reflect that older people are significant beneficiaries. It would also take pressure off the working age population who are already often financially stretched. Finally, it would draw on the significant value (despite the recent economic downturn) that older home owners have in the form of housing assets.

Appendix 1. Design issues

There are a number of design questions that need to be considered when reforming the funding of long-term care.

First, the balance between collective contributions or out-of-pocket payment by individuals for their care support. In the latter case, people simply pay for the care they need when they need it. Collective solutions involve individuals paying into a fund of some sort in advance of needing care. They are, in this case, paying to cover the risk of needing care. An example is an insurance system, but another example is where people pay general taxes for a public system that provides their care should they need it (and supposing they are eligible).

Collective contributions in the advance confer risk pooling, insurance benefits. People pay a small premium and should they be unlucky enough to develop a care need, the costs are covered. In particular, people can be covered against catastrophic' costs.

Second, a collective system can also support re-distribution goals. In this case, certain groups of people pay contributions that differ from their lifetime expected cost of care. Some people pay more and as a result other people are subsidised to pay less. A common form is re-distribution on the basis of wealth; rich people pay more (than their expected care costs) and therefore the poor pay less (or nothing). Means-tested systems have this feature. Richer people pay more in general taxation and are eligible only for limited levels of support from the public system.

Re-distribution can also happen on the basis of need i.e. low need people subsidise high need people. An example is a care system that operates eligibility thresholds, such that only people whose assessed needs exceed a given level get any public support. Low need people still pay a tax contribution but get little or no benefit.

Age is another basis for re-distribution. Contributions can be made by all ages, or limited to age groups that are likely to need care. A number of people are now suggesting a system of cohort insurance where only people 65 or over are required to make contributions to cover the costs of care for older people. In income tax-based systems all adults potentially make a contribution. In national insurance or social security systems, adults in the labour force pay.

An example of a collective system with no redistribution is private long-term care insurance. In this case premiums are adjusted by factors that correlate with the expected costs of care (e.g. age and gender; medical history; occupation etc.).

In practice, there are no (developed) long-term care systems that rely entirely on individual payments, or collective solutions without re-distribution (i.e. only private insurance). In other words, equity considerations tend to be important in long-term care, which is not surprising given the nature of needs catered for by the system.

A related point is that in almost all public collective funded systems small charges (e.g. 10% of the value of support) are levied at the point of use. These charges are often made to deter excessive or inappropriate claims to the scheme. But in theory there is no reason why they could not be dropped.

A third design question is whether or not to make collective contributions mandatory or voluntary. In regard to the latter, there tends to be a tension between a system that involves re- distribution and one that allows people to opt out. The problem is that low-risk, high wealth individuals pay more than they expect to need, and therefore see the asked-for contribution as too expensive. For example, men are likely to opt out on these grounds. An auto-enrolment system (where people have to actively opt out) offer some inertia which would slow the opt-out rate. One option would be to automatically reduce retirement pensions by the amount of the care system contribution and only restore the full pension rate if the person activity opted out. The advantage, politically, of a non-mandatory system is that the payment does not look like an additional tax.

A fourth important consideration is whether to implement a pay-as-you-go (PAYG) system or a funded system. In the former, contributions need only cover the costs of people currently needing care. In most cases, therefore, people who are paying into the system are doing so to cover the costs of someone else's care. In return they can expect other (younger or non-needy) people to be paying for their care at the time they need it, should they need support. A funded system has people make contributions in order to cover their own risk of needing care in the future. A fund is built up that would be sufficient to meet the average cost and the average risk of that person. Funded systems have the important advantage that the person is paying for their own expected care needs. But the main shortcoming is that people in the present are paying for care they might need well into the future. A 65 year old might not need any care for 20 years. By that time the costs of care will be much higher (on current trends) and the life expectancy of the 65 year old in 20 years times will be much greater than an 85 year old now. This means that funded systems cost more in the present period than PAYG systems (more than half again more expensive).

A fifth design issue concerns how closely benefits are tailored to need, and whether benefits – either as services, support or cash payments (see below) to the eligible individual – are determined by explicit criteria (e.g. in the form of a scoring algorithm) or based on the judgement and discretion of a professional (a care manager) on the ground. The latter gives more flexibility and closer tailoring of care to a person's circumstances, but accommodates more local variation (post code lotteries) and can be influenced by the nature of the relationship between care manager and service user.

Eligibility for support is usually based on an assessment of need that covers a person's impairment and functioning (both physiologically and psychologically), safety, health conditions and so on. It can also determine how much support a person is receiving from family or other informal/unpaid carers. Indeed, receipt of informal care is a very important risk factor in the need for care (Davies, Fernández et al. 2000) i.e. people need far less formal support if they are receiving informal care. There are arguments, however, that care systems should be carer blind' i.e. not account for any informal care inputs – this is a sixth design issue. On pragmatic grounds, objective and verifiable measurement of informal care inputs is difficult. More fundamentally, there are ethical reasons why people with informal care should not be penalised. And there is also a need to support informal carers themselves (although this is often achieved by means of a separate carer's assessment).

A seventh issue concerns the degree of choice afforded to the (publically-supported) service user. It is really a question of the balance of control in deciding care options as between the service user and the care manage. Consistent with the trends outlined in the introduction, many countries have introduced the option for people to take cash as an alternative to services. This pushes the balance of control to the user, but in doing so increases the administrative and transactional' burden on the individual. Cash benefit options are often chosen because they can be used to support informal care. Some countries, like Germany, offer cash benefit alternatives that have a lower value than the services offered to people, and this tends to be a very popular option (e.g. over 75% of claimants in Germany take the cash option). In part this high uptake reflects the value of being in control', but also it allows people with informal carers to pay those carers.

6. References

Audit Commission (2004). Assistive technology: Independence and well-being 4.

London, Audit Commission (www.audit- commission.gov.uk/assistivetechnology)

Barton , P., S. Bryan , et al. (2006). A National Evaluation of the Costs and

Outcomes of Intermediate Care for Older People. Birmingham and Leicester, Universities of Birmingham and Leicester. Final Report (http://www2.le.ac.uk/departments/health-sciences/extranet/research- groups/nuffield/pdf/WP114.pdf)

Baumker, T., A. Netten, et al. (2008). Costs and outcomes of an extra-care housing

scheme in Bradford. York, Joseph Rowntree Foundation

Brodsky, J., J. Habib, et al., Eds. (2003). Key policy issues in long-term care.

Geneva, World Health Organization.

Darton, R., T. Baumker, et al. (2008). Evaluation of the Extra Care Housing

Funding Initiative: Initial Report, PSSRU, University of Kent. PSSRU Discussion Paper 2506/2

Davies, B., J. Fernández, et al. (2000). Equity and Efficiency Policy in Community

Care. Aldershot, Ashgate.

Department of Health (2001). Intermediate Care. London, Department of Health.

Health Service Circular/Local Authority Circular HSC 2001/01;LAC (2001)1

Forder, J. (2002). Economic theories of purchasing. Effective purchasing for

health gain. J. Figueras, R. Robinson and E. Jakubowski, WHO.

Forder, J. (2002). Regulating entrepreneurial behaviour in social care. Regulating

Entrepreneurial Behaviour in European Health Care Systems. R. Saltman, R. Busse and E. Mossialos. Buckingham, Open University Press.

Forder, J. (2007). Self-funded social care for older people: an analysis of

eligibility, variations and future projections, Commission for Social Care Inspection & PSSRU, University of Kent. PSSRU Discussion Paper 2505 (http://www.csci.org.uk/pdf/20080128_Self- funded_social_care_for_older_people.pdf)

Forder, J., A. Netten, et al. (2007). Measuring Outcomes in Social Care: Conceptual

Development and Empirical Design, PSSRU, University of Kent. PSSRU

Discussion Paper 2422, report for Office of National Statistics

(http://www.statistics.gov.uk/about/data/methodology/specific/PublicS

ector/output/qmf/downloads/PSSRU_QMF_Interim_Report_Oct_2007.pdf) Forder, J., R. Robinson, et al. (2005). Theoretical perspectives on purchasing.

Effective Purchasing for Health Gain. J. Figueras, R. Robinson and E.

Jakubowski. Buckingham, Open University Press.

Glendinning, C., D. Challis, et al. (2008). Evaluation of the Individual Budgets Pilot

Programme, University of York , London School of Economics, University of Kent (http://www.pssru.ac.uk/pdf/IBSEN.pdf)

Glendinning, C. and E. Newbronner (2008). "The effectiveness of home care

reablement - developing the evidence base." Journal of Integrated Care 16(4): 32-39.

Godfrey, M., J. Keen, et al. (2005). An Evaluation of Intermediate Care for Older

People, Institute of Health Sciences and Public Research, University of Leeds

Godfrey, M., C. Surr, et al. (2005). Prevention and Service Provision: Mental

health problems in later life. Leeds and Bradford, Institute of Health Sciences and Public Health Research, University of Leeds and Division of Dementia Studies, University of Bradford. Study commissioned by Age Concern and the Mental Health Foundation

King, N. (2001). "Extra care housing: Capital and revenue funding." Housing Care

and Support 4(4): 24-28.

Knapp, M. (1984). The economics of social care. London, Macmillan.

Knapp, M., B. Hardy, et al. (2001). "Commissioning for quality: Ten years of social

care markets in England." Journal of Social Policy 30(2): 283-306.

Melis, R. J. F., M. G. M. Olde Rikkert, et al. (2004). "What is intermediate care?"

BMJ 329(7462): 360-361.

Netten, A., R. Darton, et al. (2008). Evaluation of the Extra Care Housing Funding

Initiative: Summary of Initial Findings. Canterbury, PSSRU, University of Kent. PSSRU Research Summary 47 (http://www.pssru.ac.uk/pdf/rs047.pdf)

Netten, A. and B. Davies (1990). "The Social Production of Welfare and

Consumption of Social Services." 10(3): 331-347.

Norton, E. (2000). Long-term care. Handbook of Health Economics. A. Culyer and

J. Newhouse. Amsterdam, Elsevier Science B.V. 1.

OECD (2005). Long-term Care for Older People. Paris, OECD Publishing.

Oxera (2007). What is the economic impact of Jersey's aging population? Oxford,

Oxera Consulting Ltd

Pilkington, G. (2008). "Homecare re-ablement: Why and how providers and

commissioners can implement a service." Journal of Care Services Management 2(4): 354 - 367

Poole, T. (2006). Funding Options for Older People's Social Care. London, King's

Fund. Wanless Social Care Review Background Paper 12 (http://www.kingsfund.org.uk/document.rm?id=6330)

Poole, T. (2006). Telecare and Older People. London, King's Fund

(http://www.kingsfund.org.uk/document.rm?id=6361)

The Information Centre (2008) Community Care Statistics 2007: Home care

services for adults, England. London & Leeds, The Information Centre, Government Statistical Service (http://www.ic.nhs.uk/webfiles/publications/Home%20Care%20%28HH 1%29%202007/HH1%202007.pdf)

Wanless, D., J. Forder, et al. (2006). Securing Good Care for Older People: Taking a

long term view. London, King's Fund

Wistow, G., M. Knapp, et al. (1994). Social Care in a Mixed Economy. Buckingham,

Open University Press.

Wistow, G., M. Knapp, et al. (1996). Social Care Markets: Progress and Prospects.

Buckingham, Open University Press.