Future costs for long-term care: Cost projections for long-term care for older people in the United Kingdom
Introduction
The subject of long-term care (LTC) is receiving increasing attention both in the research community and by various countries’ governments due to the belief that an ageing population will greatly swell the demand for long-term care services and create huge public expense. One of the issues which needs to be determined is by how much demand will increase; another is to address the ambiguity over whether long-term care is a response to a medical condition, a social need or both. The corollary is to decide how the burden is to be shared between the individual, the family and the state.
LTC is administered to people who have reached a stage in life in which they are dependent on others for social, personal and medical needs. It is usually associated with the very old but, in fact, could begin at any age depending on the reasons for the disability—perhaps a road accident, a mental or a congenital condition. For some, long-term care may be needed over an extended period whereas, for others, it is required in the period immediately prior to death. In this paper the focus is on older people since they tend to have the greatest need for long-term care.
The purpose of this paper is to analyse the sustainability of the UK system for provision of long-term care in the light of the changes in demography and health status among older people that are expected in the future. In particular, we wish to find out how demand for LTC will evolve and to what extent there will be sufficient supply to meet demand. In terms of formal care, this requires an estimate of how much the public purse, and hence the taxpayers, will be burdened with LTC costs in the future. As far as informal care is concerned, it involves estimating whether there will be enough carers under the assumption that current patterns of provision do not change in the future.
Whilst slightly different LTC systems operate within Scotland and the rest of the UK, the basic premise is that, in the UK, LTC relies less on public financing than most other European systems. Eligibility to free or subsidised care is based on means testing, and under some circumstances, the value of a person's home into account as well as income and other assets. In the UK, the LTC sector is characterised by dual arrangements, in which different principles apply to health care services—provided by the National Health Service (NHS) and social services—financed by local authorities.
Local authorities have two main sources of funding—government grants and locally raised revenue in the form of council tax. The funds are not earmarked, but there are recommendations and management targets on how to spend the money and the service levels expected. By contrast, the NHS is responsible for funding some nursing home places and also finances nursing care in all care settings based on health related criteria. The NHS is financed by the constituent countries of the UK and is largely financed out of income taxation [1]. There is only a small market for private long-term care insurance, and up to the end of the year 2000, fewer than 40,000 policies had been sold [2].
As far as this research is concerned, the basic demographic trends are taken as ‘given’, but it is important to understand how these trends have emerged in general terms and what the implications are in terms of the demand for long-term care. An ageing population is a trend common to all developed countries that manifests itself in terms of an increasing proportion of older people in the population. This has arisen not only because people are living longer but also because women are having fewer children than in the past. In several countries, the population has stagnated or is set to decline (the US being a notable exception).
Over the past 25 years, there has been an intense academic debate on the implications for healthy life expectancy (HLE) of falling mortality rates. Three competing hypotheses have been proposed. The most optimistic one, suggesting a compression of morbidity, is due to Fries [3]. According to this perspective, adult life expectancy is approaching its biological limit so that if disability spells can be postponed to higher ages the result will be an overall reduction in the time spent disabled. By contrast, Gruenberg [4] suggested an expansion of morbidity based on the argument that the observed decline in mortality was mainly due to falling accident rates. The third hypothesis was proposed by Manton [5], according to whom the development in mortality and morbidity is a combination of the two, which could lead to an expansion of the time spent in good health as well as the time spent in disability.1
There is, however, not yet enough empirical evidence available to draw a definite conclusion on how the gap between healthy life expectancy and total life expectancy is behaving in all countries. According to national statistics for Great Britain, HLE at age 65 increased by around 1.5 years between 1981 and 2001, whereas overall life expectancy increased by 2.1 years for women and 3 years for men, suggesting that the gap has expanded in the last 20 years. These figures are consistent with Manton's argument above [6].
The key implication is that trends in health could make a significant difference to costs and therefore public policy. We therefore need to ensure that our analysis takes into account a range of possible health scenarios. For this part of the analysis we rely on previous work by Rickayzen and Walsh [7] who developed a methodology for projecting disability prevalence rates, allowing for health trends (see Section 2.1).
One of the first rigorous reports on the future costs of long-term care was provided by Nutall et al. [8]. The projection was based on a multi-state model of disability, where the three states are assumed to be healthy, disabled and dead. Separate series of models were built to incorporate severity of disability in which no recovery was allowed once the particular disabled state has been reached. The OPCS study of disability provided the basis for prevalence rates (with the implicit assumption that prevalence rates by age had remained constant between 1986 and 1991, the base year). The study projected a rapid increase in the demand for long-term care from 2011 onwards. In order to estimate the future costs of LTC, it was assumed that LTC costs remain constant in terms of GNP (alternative scenarios with changing relative prices were also considered). According to the central projection, LTC costs as a share of GNP would increase by 47% (from 7.3 to 10.8%).
A more recent projection has been provided by the PSSRU [9] (see [10] for the most recent update) with the Personal Social Services Research Unit (PSSRU) model. The PSSRU model, originally developed for the Royal Commission on Long-term Care [11], assumes that dependency rates by age and sex remain constant over the projection period (ending at 2031) and uses a cell-based model to project the future demand for LTC services and the implied costs. The dependency measure used in the PSSRU model is the number of activities of daily living (ADLs) and instrumental activities of daily living (IADLs) failed by the individual, which are based on typical daily activities such as cooking. The outcome of the baseline scenario of the model is that formal LTC service will have to expand by 61% between 1995 and 2031. Further attempts to model future LTC costs have been made by London Economics and the Institute for Public Policy Research [12], and the Department of Health [13].
The present study differs from the PSSRU model in several ways. Firstly, the models differ in the definition of dependency. The PSSRU model uses ADLs and IADLs for the non-institutionalised population and treats institutionalisation as a distinct kind of dependency. Our model, on the other hand, uses the wider OPCS scale and takes accounts of the heterogeneity—in terms of dependency—of the institutionalised population. Secondly, the PSSRU model makes projections for England, whereas we are concerned with the entire UK. Thirdly, the definition of LTC is different; we use a narrower definition of LTC, covering only institutional care and certain home care services, whereas the PSSRU model also covers care settings such as long-stay hospital care, day care and community nursing. Finally, the basis of funding is different in the two models; we use the labour remuneration of the working population as a basis, whereas the PSSRU model assumes a constant growth in GDP. A comparison of the results achieved by us and by Wittenberg et al. [10] is provided in Appendix A.
There are two main advantages to our approach. Firstly, by relying on transition probabilities as the basis for projecting future needs, and not simply a demographic extrapolation of current needs, we are likely to get a more accurate estimate of the levels of future needs, as well as in the range in uncertainty which we need to consider. Secondly, we avoid having to take the detour of first calculating costs of LTC and then comparing them to the GDP, which has been projected to grow at some constant rate (as in the PSSRU model). On the contrary, we acknowledge the fact that the capacity of the economy itself, especially where a labour-intensive service such as LTC is concerned, depends mainly on the size and structure of the labour force. It is unlikely that productivity increases in the economy will lessen the burden of LTC financing, and hence we use as our baseline scenario an assumption that prices of LTC services increase in line with general earnings.
This increased accuracy concerning the prevalence of dependency and the overall macro-economy, comes at the cost of less flexibility in other parts of the model. Hence, we are not able at this stage to model the implications of changing family structures for the formal care sector, or the implications of shifts in the income distribution of older people for public finances. Thus, a crucial assumption underlying our work is that trends in dependency and demography are the main drivers of LTC expenditure.
The paper is organised as follows. In Section 2, the different elements of our projection model are presented in more detail. In Section 3, we present results and undertake a sensitivity analysis. Section 4 concludes. Appendix A provides a comparison of our results with the PSSRU [10] model and in Appendix B we analyse the effects of altering the assumption with regard to the relative price of care services.
Section snippets
Projection model
Our projection model consists of several different components. An overview is given in Fig. 1 which will be explained in more detail in this section. From our projections, we derive two kinds of results; firstly, an estimate of the future costs of LTC to the public purse, expressed as a proportional income tax and, secondly, an estimate of the future surplus or shortfall of the number of informal carers relative to the demand for informal care. In Fig. 1, arrows going downwards represent
Results
The results are presented below, in the same order as the model and its assumptions were outlined in Section 2 (cf. Fig. 1).
Conclusion
Long-term care is a very complex issue and the development of demand for LTC services is determined by, inter-alia, the prevalence of disability in the population, economic factors, the institutional environment, preferences, family structures, and the interaction between them. In order to project the future development of the LTC sector, it is necessary to focus on the aspects that are deemed particularly important. In this paper, we have focused mainly on how the prevalence of disability
References (28)
- et al.
United Kingdom health care systems in transition
(1999) - et al.
Long term care
(2001) Aging natural death and the compression of morbidity
New England Journal of Medicine
(1980)The failures of success
Milbank Memorial Foundation Quarterly/Health and Society
(1977)Response to an introduction to the compression of morbidity
Gerontologica Perspecta
(1987)Healthy life expectancy at birth and at 65 in Great Britain and England, 1981–2001
(2002)- et al.
A multi-state model of disability for the UK: implications for need for long term care for the elderly
British Actuarial Journal
(2002) - et al.
Financing long-term care in Great Britain
Journal of the Institute of Actuaries
(1994) - et al.
Demand for long-term care: projections of long-term care finance for elderly people
(1998) - Wittenberg R, Pickard L, Comas-Herrera A, Davies B, Darton R. Future demand for long-term care, 2001 to 2031....