3.1 Introduction

The demand for long-term care is expected to rise dramatically over the next decades due to rapid population ageing. While most high-income countries have developed a system of long-term care to respond to increased demand, there are important variations in their approach to defining eligibility, coverage and equity. In low and middle income countries (LMICs), on the other hand, long-term care continues to be a largely neglected or low priority policy issue (Lloyd-Sherlock, Reference Lloyd-Sherlock2014). LMICs will experience fast population ageing and will face a steep increase in the demand for long-term care in the next decades, which will require effective policy responses (OECD, 2019). Establishing a long-term care system for older people should therefore be a major policy priority for countries across the globe. An important number of older people will not be able to perform basic ADLs and will need support with them, such as getting in and out of bed, getting dressed, going shopping, doing laundry, and going out to meet friends. They will also require support with more complex tasks including managing medical care for multiple, chronic conditions.

As explained elsewhere in this volume, the first and most common source of long-term care support for older people worldwide is commonly known as informal care, typically provided by a spouse or adult child, or in some cases, a friend or neighbour (Colombo & Mercier, Reference Colombo and Mercier2012). However, informal care may not be always available or may not be the preferred option. This might happen, for example, when needs are – or have become – too severe and require more intensive long-term care; when informal carers are not willing and/or able to provide support, often due to work-related time or geographical constraints; or when the person in need simply does not have an adequate social network from which to seek support (Carrino et al., Reference Carrino, Nafilyan and Avendano2022). In these cases, formal care may be necessary, that is, care provided by professional carers such as nurses, personal carers and personal assistants. While in principle people seeking formal long-term care have the option to purchase services from private providers, the costs of private formal long-term care of an acceptable quality can prove unaffordable for people in high-income countries, and prohibitive for people in LMICs, especially if support is required for continued, recurrent or extended periods. An alternative strategy would be to purchase long-term care insurance earlier on in life but, for reasons explored in chapter 4, both demand and supply of private insurance is still very limited in most countries.

Due to the fact that most older individuals will not be able to cover the full costs of long-term care, or lack insurance for long-term care costs, many countries have established local public institutions and services that are in charge of financing, organising and/or providing, at least to some extent, long-term care services. However, people seeking public support for formal long-term care will often find the process more complex than for health care. While the exact procedures differ from country to country, most systems involve a set of needs assessment tools and eligibility rules for publicly financed and/or publicly provided long-term care services. As this chapter shows, procedures can be intricate, may differ from region to region within a country, and typically involve many stakeholders as well as numerous rules with sometimes complicated calculations. People in need of long-term care and their relatives and friends may find it difficult to predict what services they are eligible for and how much these services will cost them. Moreover, as with private provision, local supply of public formal long-term care may also be constrained, for example due to shortages in skilled workers and/or low public spending.

In this chapter, we focus on the challenges that arise with the measurement of long-term care needs to determine eligibility for long-term care, as well as on the outcomes of eligibility rules on access to care and wellbeing. We focus our analysis on the case of European countries with well-established long-term care systems as they have extensive experience with the use of eligibility rules, but our aim is to draw lessons from these countries for other countries around the globe in the process of setting up long-term care systems.

Eligibility rules impact how much and what type of care individuals receive. They are useful tools to identify persons with the greatest need for care, and to ensure resources are distributed in a way that best addresses equity. And yet, we document that defining eligibility for long-term care is fraught with challenges arising from the lack of a universally agreed approach to measuring the concepts that define health and social need for long-term care, and difficulties in linking different levels of need to the right amount and type of long-term care. As a result of the design of eligibility rules, some people with functional or cognitive limitations may either be ineligible for publicly financed and/or publicly provided long-term care, or, despite being eligible, receive an amount of support that is insufficient, and/or face additional unaffordable OOP costs of care. This may lead them to rely exclusively on informal care when this may not have been their preferred option, or go without care altogether. Living with unmet needs not only reduces their ability to live independently but may also put them at higher risk of avoidable and costly hospital care. Systems may therefore develop eligibility rules that are consistent with the objectives of the long-term care system to improve coverage and equity, while at the same time maintaining efficiency.

We argue that policy makers should consider expanding eligibility rules for long-term care. We show that existing eligibility rules are an effective policy tool to increase older people’s access to care, although they are less effective for lower socioeconomic groups (e.g., lower educated groups). We also show that these rules reduce the risk of extreme poverty resulting from care costs and they improve wellbeing and quality of life.

Finally, we argue that governments around the world ought to reflect carefully about their criteria to define eligibility rules, as they are crucial to achieving key goals of the system such as universal long-term care coverage, effectiveness, people-centredness, accessibility and financial sustainability.

This chapter is organised as follows. In section 3.2 we introduce the policy challenge of defining need for long-term care. Section 3.3 then explores how countries in Europe undertake assessment of long-term care needs, how they differ and what the implications are for individuals in need of care. Section 3.4 then provides an examination of the complexities of measuring need for long-term care, introducing concepts of disability, dependency, frailty and independence, exploring how these concepts are operationalised in needs assessments. Section 3.5 goes on to look at the question of whether the availability of informal care should be taken into account in needs assessments. Section 3.6 examines the effects of different eligibility rules on rates of coverage in a population, while section 3.7 looks at their implications for the intensity of coverage. Section 3.8 explores the role of means testing in determining eligibility. Section 3.9 discusses the implications of eligibility rules for utilisation of long-term care and the wellbeing of older people. In concluding, section 3.10 draws the arguments together, highlighting key takeaways for policy makers based on our analysis.

3.2 Defining ‘need for long-term care’ is a policy challenge

A key step in determining eligibility for public formal long-term care is establishing the need for care. While health policy is often concerned with the question of how to deliver care based on need as determined by a medical diagnosis, the definition of need for long-term care rests on multiple contested concepts such as frailty, disability, dependence and autonomy (Brugiavini et al., Reference Brugiavini, Carrino, Orso and Pasini2017; Colombo & Mercier, Reference Colombo and Mercier2012). The definition of these concepts is the subject of decades of debates in clinical gerontology, geriatrics, nursing, biodemography and sociology. Operationalising these concepts raises challenging questions for societies and governments: can populations agree on a list of tasks defined as basic or essential to maintain independent living and human dignity, and if so, how can these tasks be measured? Should an individual’s self-reported ability to perform tasks be considered among the criteria, or should other external assessments (e.g., physical performance) be required to establish a person’s ability to perform tasks? Should factors other than needs – such as financial means or availability of informal care – be considered when determining eligibility, especially in the context of tight fiscal conditions in most countries around the world? What other factors should be considered, given that governments have many competing objectives, including efficiency, equity and quality?

The question of who should be eligible for long-term care touches on critical questions about the role of the state when individuals need support with their functioning in daily life. For example, in Nordic European countries, advanced social insurance systems confer an important role on the state in financing, organising and/or providing long-term care. By contrast, in the United States, long-term care is often the responsibility of individuals and their families, except for cases of extreme poverty where Medicaid may step in to cover the costs of long-term care. When establishing the rules for eligibility, governments may have in mind specific objectives and values that respond to different concepts of the role of the state. The WHO health system framework states that a health policy should aim to ensure universal health coverage, access and financial protection (WHO, 2007a), while the OECD health framework highlights the importance of effectiveness, people-centredness, accessibility and financial sustainability of health systems (Carinci et al., Reference Carinci, Van Gool, Mainz, Veillard, Pichora and Januel2015).

In achieving these objectives, countries designing long-term care systems may need to take into account three important dimensions: efficiency in the use of limited resources; distribution according to some principle of equity or social justice; and preservation of individual freedom (Barr, Reference Barr and Barr2012). Although most countries may agree on the importance of all three aims, they differ in the definition and weight they assign to each of them, resulting in significant variations in long-term policy. Underlying these variations are deep political and ideological debates. From a utilitarian perspective, for example, the aim of long-term care policy may be to maximise total welfare (for older people, their carers and society as a whole). In contrast, it has been increasingly argued that freedom of choice is essential for any long-term care policy, and many countries have moved towards providing more choice in the type of care received, for example through the use of cash for care programmes (Gori, Reference Gori2012). The definition and weight given to each of these aims imply a normative judgement, one that requires substantial political debate.

The definition and quantification of need ultimately have important implications for one of the key tasks of any long-term system: establishing eligibility for publicly funded long-term care services. Most social benefits are dependent on well-defined criteria to define eligibility, such as a specific age threshold (e.g., reaching the age of 65 for pensions), being below a certain level of income (e.g., for social assistance programmes), or having contributed to an insurance programme for a defined period and experienced a well-defined shock (e.g., unemployment insurance). On the one hand, eligibility criteria are a mechanism by which governments regulate the number of beneficiaries, control costs and plan for the resources required to run the system (Buscher et al., Reference Buscher, Wingenfeld and Schaeffer2011). On the other hand, eligibility criteria are also deeply rooted in complex concepts that signal whether individuals need government intervention. In this chapter we will argue that eligibility criteria for long-term care have an impact on coverage and accessibility, but they are also inherently linked to concepts related to functional capacity, the ability to live independently, and the potential to prevent decline or compensate for the loss of function.

In the context of limited resources, several factors come into play to shape the way governments determine eligibility for publicly funded long-term care. First, the definition and quantification of need establishes whether and to what extent individuals face limitations in functioning. However, not all individuals in need will be automatically eligible for public support, and individuals with different levels of need may be assigned different levels and types of care. Typically, governments define thresholds based on some consensually agreed level of need above which individuals are considered to face difficulties in functioning independently.

Second, some governments may consider that need itself is not sufficient and incorporate some assessment of the ability to pay in the form of means testing based on income and/or wealth. By targeting those on the lowest incomes, governments may wish to address equity objectives under the assumption that those with enough resources will be able to purchase care in the market. Governments would therefore provide support for essential care tasks, leaving individuals to co-pay the costs of board and lodging.

Third, in defining eligibility for long-term care, a few countries consider whether individuals in need have a partner or adult child who can support them. The assumption is that family members will perform at least some of the care for their older relatives.

These differences result in varying models of long-term care provision and financing. Recent classifications (see, for example, Fernández et al., Reference Fernández, Forder, Trukeschitz, Rokosová and McDaid2009; Gori and Fernández, Reference García-Gómez, Hernández-Quevedo, Jiménez-Rubio and Oliva-Moreno2015; and Barber et al., Reference Barber, Ong, Han and Haring2020) distinguish between countries that minimise state intervention and provide support only to the lowest-income subgroups in the population (safety net systems, such as in Australia, England and the United States); countries with a universal approach that aims to provide equitable access based on health needs (e.g., Denmark, Japan, the Netherlands), and mixed systems which combine universal entitlements with a means-tested approach (e.g., Austria, Belgium, France). As we shall see, eligibility rules and their outcomes are not necessarily homogeneous even within these typologies.

Finally, governments face a trade-off between the transparency of long-term care eligibility rules and their effectiveness in matching resources to needs, as thoroughly discussed by Fernández et al. (Reference Fernández, Forder, Trukeschitz, Rokosová and McDaid2009). On the one hand, transparent and easily interpretable rules are desirable as they facilitate individuals’ care allocation planning, as well as public auditing. Long-term care systems can be broadly split between those adopting algorithm-based eligibility rules, and those adopting a case-managed approach to determine eligibility. Algorithm-based systems are more transparent, as they include an explicit list of measurable tasks and thresholds that characterise the eligibility score (examples are France, Germany, Japan, Spain). Case-managed systems determine eligibility on a case-by-case basis: front-line workers (the case managers) exercise significant discretion in determining individuals’ entitlement to long-term care, and the evaluation of need most often includes individual or contextual characteristics not easily measurable, such as the quality of the informal care support (examples are England, Denmark, Sweden). Case managers often follow an official guideline which does not include an explicit threshold or algorithm. Case-managed systems are therefore deemed less transparent and more complex. On the other hand, a long-term care system is more effective in matching care packages to individuals’ needs if it allows for an assessment of needs that accounts for the complex interaction of the applicant’s multiple circumstances, such as, for example, physical and mental wellbeing, attitudes, income, family context and support network. An algorithm-based system, although more transparent, is less effective in matching resources to needs than case-managed systems, which are more flexible and comprehensive in the evaluation of individual circumstances.

3.3 Assessment of needs and eligibility rules: how to do it and what are the consequences?

In most OECD and EU countries, a resident can apply for public long-term care services and support, but geographical coverage is not always comprehensive, coverage may not be ensured in all cases, and older people, their relatives and their advocates might find the application processes – with multiple eligibility criteria, numerous stakeholders and intricate rules – difficult to navigate.

In countries with well-established long-term care systems, eligibility is typically determined through so-called long-term care needs assessments, which are used to determine whether an older person is eligible for public support and services. This would ideally enhance governments’ ability to maximise access, affordability, equity (i.e. the most vulnerable are adequately protected) and efficiency (i.e. the desired level of results is achieved at the lowest cost to the public purse). One way in which needs assessments do this is by determining eligibility based on a minimum level of needs, and by associating higher intensity of care with higher severity of needs. The exact procedures employed in different countries are very heterogeneous, as detailed in recent reviews such as Gori and Fernández (Reference García-Gómez, Hernández-Quevedo, Jiménez-Rubio and Oliva-Moreno2015), Oliveira Hashiguchi and Llena-Nozal (Reference Hashiguchi and Llena-Nozal2020) and Brugiavini et al. (Reference Brugiavini, Carrino, Orso and Pasini2017). As already discussed, eligibility rules differ, for example, depending on who conducts the assessment, which types of limitations and needs are included, the way different limitations are scored to determine overall severity of needs, whether social structures (e.g., adult children) are considered, and finally how the results from needs assessments determine eligibility for specific levels of long-term care support and services (Table 3.5 in the chapter appendix provides an overview of needs assessments in different countries).

The types of limitations and difficulties that are considered relevant in needs assessments differ from country to country. In what follows, we will mostly concentrate on countries that adopt algorithm-based eligibility rules, where the process of assessment of need is explicitly defined in the legislation. A frequent element across countries’ rules is the focus on ADLs, with many programmes also including IADLs (introduced in chapter 2). However, as shown in Brugiavini et al. (Reference Brugiavini, Carrino, Orso and Pasini2017), the way in which such difficulties are operationalised may vary greatly: for example, several countries do not include all IADLs (such as in major programmes in Italy and France, whereas in Germany IADLs are included in the assessment but not in the eligibility algorithm), and others do not include all ADL tasks (e.g., in Italy and England). Moreover, many programmes give equal weight to each ADL and IADL limitation (e.g., in Belgium, Czech Republic, England and Poland). However, existing theories in gerontology suggest that IADLs and ADLs can be interpreted within a hierarchical structure (Njegovan et al., Reference Njegovan, Man-Son-Hing, Mitchell and Molnar2001). In general, programmes that give more weight to prevention typically give a relatively higher weight to limitations in IADLs and mobility (e.g., Austria), as opposed to being focused mainly on ADL limitations (e.g., Poland), because IADLs and mobility represent more complex tasks, which tend to emerge earlier than the more basic ADL tasks.

Some assessments also take into account social activities and factors such as loneliness, cognitive status and mental wellbeing. For example, needs assessments in Belgium incorporate awareness of dangers and the ability to have a social life, while in Denmark, assessments consider the scope for rehabilitative treatment to restore functional abilities. Several countries have updated needs assessments to include not only functional limitations, like walking, but also mental and cognitive impairments. In Germany, for example, needs assessments recognise that dementia requires a different kind of support. While dementia patients may not exhibit physical impairments, they may face difficulties performing ADLs autonomously.

There is significant heterogeneity between countries in the way needs assessments aggregate different limitations to produce a single measure of overall severity of long-term care needs, which is used as a benchmark for determining someone’s eligibility status. The reasons for these differences are not well understood, but are likely to involve a combination of: differences in the relative weight each system gives to concepts such as frailty and dependency; the relationship between the ministries of health and social affairs; available budgets; the supply of informal care; cultural factors (e.g., family as first line of support instead of the state); and the infrastructure and built environment, among other factors. In some countries, scores are calculated using different weights for different limitations, and needs assessments may require that the older person suffer from specific limitations to be eligible for any kind of public support. In Germany, a category called ‘self-sufficiency’, which is composed of thirteen indicators including eating and drinking, has the highest weight, while ‘mobility’ has the lowest weights. In Austria, no government support is possible if care needs amount to fewer than 65 hours of care per month. Some countries group scores into categories, which are then associated with levels of support and specific services. Categories with fewer limitations may be ineligible for public support. Countries differ in how many categories they set, how they combine scores into categories, and how categories are associated with specific care services. While some countries have three to five categories, Luxembourg sets fifteen categories based on the amount of time allocated for care per week. Moreover, the weight assigned to each ADL or IADL difficulty is typically different across programmes: some rules treat each ADL or IADL equally (e.g., in the Czech Republic), while others postulate that the impact of a loss of autonomy differs depending on the specific limitations (e.g., in Austria). For example, difficulties in eating might be assumed to be more relevant than difficulties in dressing or in doing housework.

Not all countries have standardised long-term care needs assessments. Austria, Germany and Luxembourg have nationwide assessments using the same instruments, while Italy complements a national programme with regional ones whose criteria are different across territories. France uses a standardised assessment tool nationwide but there seems to be much variation across regions in the interpretation of the tool, generating differences in the number of beneficiaries. In Estonia, different agencies have different assessment scales and procedures. In Belgium, Canada and the United States, there are both federal and regional, state and municipal benefits and schemes, with their own specific rules and regulations. In Portugal, there is one needs assessment scale for cash benefits and another for in-kind care, the latter being left to the discretion of the private care provider. In Nordic countries (Denmark, Iceland, Finland, Norway, Sweden), municipalities are responsible for conducting needs assessments and setting eligibility criteria on a case-managed basis, although Finland is rolling out a nation-wide standardised needs assessment scale.

Countries and subnational areas differ as well in terms of the staff involved in long-term care needs assessments. A number of countries use multidisciplinary teams to perform assessments and the composition of the teams varies across countries. In Belgium and Finland, multidisciplinary teams conduct needs assessments. In Latvia, the teams include general practitioners (GPs) and social workers, while in Germany specialised doctors and nurses are involved. In Austria and England, assessments are based on a physician’s evaluation. In California (United States), a social worker performs the assessment. In Japan, a computer-aided initial assessment (based on an 85-item questionnaire) is used to assign each applicant to one of seven levels of long-term care need. The Japanese Care Needs Certification Board, a committee composed of medical and other professionals, then reviews the results from the initial assessment.

In many countries, long-term care rules and benefits are decentralised. For example, the Austrian long-term care system consists of three pillars. The first pillar is a cash benefit, the second one consists of measures to support caregiving relatives, and the third pillar consists of in-kind benefits. The first two pillars are the responsibility of the federal government, while the federal states are responsible for providing the third pillar. In Belgium, competencies are shared between the federal government, the communities (the Dutch-speaking or Flemish community, the French-speaking or Walloon community and the German-speaking community) and the Brussels Capital region. The Belgian long-term care system consists of ‘a mix of different services and measures, funded through different sources and organised at different levels’ (Pacolet & De Wispelaere, Reference Pacolet and De Wispelaere2018:5). In Estonia, social welfare services are organised at both the state and municipality level, and personal care services are mostly organised by the local government (Paat-Ahi & Masso, Reference Paat-Ahi and Masso2018). In Italy, the public long-term care system is organised around two pillars (Matteo et al., Reference Matteo, Emmanuele, Michele and Natili2018): first, the companion allowance, which is a cash benefit run centrally by the National Institute of Social Security; and second, home and residential care services provided by municipalities (personal care) and regions (health and nursing care).

Finally, eligibility for public long-term care support may also depend on social networks and external factors, such as the availability of an informal carer (this is the case in Croatia, Italy, the Netherlands and Ireland). We will discuss this in section 3.5.

By the age of 65, most individuals will have accessed the health care system multiple times throughout their lives. They are likely to know enough about the health care system to be able to understand and anticipate what type of needs merits which type of care, where to access it, and whether they will be able to afford it. At the very least, they know where to find help navigating the system, typically through their local GP or family doctor. This is not necessarily true for the long-term care system, which most 65-year-olds may have never accessed before and, of which they may have a limited understanding (Bottery et al., Reference Bottery, Varrow, Thorlby and Wellings2018). Even identifying where to seek help may be challenging. Even more challenging is determining how much care will cost them out of pocket, given the very intricate rules and calculations that are used in many countries (see annex E in Hashiguchi & Llena-Nozal, Reference Hashiguchi and Llena-Nozal2020). While there is limited research into how well populations understand the eligibility requirements and OOP costs of long-term care, a cursory look at the rules in some countries suggests some older people might not be able to predict their eligibility and costs.

3.4 The challenge of defining and measuring long-term care need and eligibility for care

Measures of functioning are generally stronger predictors of outcomes in older age than single diseases. However, health surveillance systems are often developed around specific diseases such as cancer or cardiovascular disease, while there is no system focused on collecting information on functioning. This is partly due to disagreements on the definitions of long-term care need and instruments. Specific instruments do exist to measure specific dimensions of functioning, such as physical or cognitive function, but these instruments are often used for specific research or clinical purposes. Part of the challenge arises from the difficulty with defining key concepts, particularly dependency and disability in older age.

The International Classification of Functioning, Disability and Health (ICF) (WHO, 2007b) and the UN Convention on the Rights of Persons with Disabilities (UN, 2006) offer an ‘individualistic’ definition of dependency, ratified by the European Union (Becker, Reference Becker, Becker and Reinhard2018). The UN definition states that ‘persons with disabilities include those who have long-term physical, mental, intellectual or sensory impairments which in interaction with various barriers may hinder their full and effective participation in society on an equal basis with others’ (UN, 2006: Article 1). This definition emphasises impairments in individual function as the key element determining whether older individuals experience disability and dependency. In line with this concept, the assessment of dependency is often based on one of a number of scales that cover various dimensions of capacity or incapacity.

Many instruments are based on ADL scales. The rationale is that individuals unable to perform key everyday activities on their own(e.g.,walking, getting out of bed, bathing) are considered to be dependent. These scales identify both whether individuals may be in need of care and the degree of dependency, by measuring and weighting the number of activities with which individuals experience difficulties, as well as the level of difficultly experienced. The expectation is that these scales provide a systematic assessment that allows comparisons across individuals and over time. Disability is primarily diagnosed based on self-reports of difficulties with tasks, but performance-based tests to measure function are increasingly available in clinical and research settings. The most common scales include assessment of disability in self-care tasks (ADLs), household management (IADLs) and risk of mobility limitations (Chaves et al., Reference Chaves, Garrett and Fried2000; Fried et al., Reference Fried, Ferrucci, Darer, Williamson and Anderson2004).

Although each country’s framework applies a somewhat different definition (e.g., some systems emphasise deficiencies in performing common household chores, while other systems emphasise limitations with physical or psychological functioning), all systems imply a definition of dependency that hinges on the loss of autonomy to perform certain activities of daily life. The assessment of whether a person is dependent on long-term care is typically made by trained personnel, and it is used for decisions regarding the provision of long-term care and other social benefits. In some countries, such as the Nordic European countries, the assessment is made to measure the level of dependency and to establish the amount and specific type of care that individuals may require (Becker, Reference Becker, Becker and Reinhard2018). In other countries, the assessment of dependency may be also be used for classifications of individuals based on a detailed point system, on the basis of which eligibility for cash benefits is established.

A critique of the focus of long-term systems on disability and dependency is that these concepts do not reflect the full spectrum and trajectories that characterise the process of ageing. In response, frailty is also often evoked as an important and distinct concept in understanding ageing and resilience in older age.

WHO defines frailty as ‘a clinically recognisable state in older people who have increased vulnerability, resulting from age-associated declines in physiological reserve and function across multiple organ systems, such that the ability to cope with everyday or acute stressors is compromised’ (WHO, 2017:VII). Ageing individuals move along a continuum that stretches from robustness to pre-frailty, frailty and care dependence. Frailty is thus a different concept than disability and may be critical to identifying opportunities for prevention and care.

Two major models to define frailty exist: the frailty phenotype, also referred to as Fried et al. (Fried et al., Reference Fried, Cohen, Xue, Walston, Bandeen-Roche and Varadhan2021; Fried et al., Reference Fried, Ferrucci, Darer, Williamson and Anderson2004), and the frailty index (WHO, 2017). The frailty phenotype is defined as a clinical syndrome and pathophysiology with validated clinical presentation (Bandeen-Roche et al., Reference Bandeen-Roche, Xue, Ferrucci, Walston, Guralnik and Chaves2006; Fried et al., Reference Fried, Tangen, Walston, Newman, Hirsch and Gottdiener2001), often referred to as physical frailty (Fried et al., Reference Fried, Cohen, Xue, Walston, Bandeen-Roche and Varadhan2021). Based on a deficit accumulation model, the frailty index model emphasises the presence of multiple clinically identified diseases, their clinical manifestations and consequences, as well as various risk factors (Mitnitski et al., Reference Mitnitski, Mogilner and Rockwood2001; Rockwood & Mitnitski, Reference Rockwood and Mitnitski2007). Both measures are strong predictors of mortality and institutionalisation, but they are based on different theories on underlying aetiologies, and use different measures to capture different processes. Other definitions integrate cognitive function and emotional or psychosocial dimensions, although a common critique is that these constructs are distinct from phenotypic frailty (Fried et al., Reference Fried, Cohen, Xue, Walston, Bandeen-Roche and Varadhan2021).

Physical frailty is a distinctive high-risk clinical state that signals a decrease in reserve and high vulnerability to stressors. Underlying the concept of frailty is the notion that a healthy organism is composed of systems with independent functions, each of which deteriorates with age (Fried et al., Reference Fried, Cohen, Xue, Walston, Bandeen-Roche and Varadhan2021). Physical frailty is typically diagnosed based on three or more of five key clinical signs and symptoms: physical weakness, slow walking speed, low physical activity, fatigue or exhaustion, and unintentional weight loss (Bandeen-Roche et al., Reference Bandeen-Roche, Xue, Ferrucci, Walston, Guralnik and Chaves2006; Fried et al., Reference Fried, Cohen, Xue, Walston, Bandeen-Roche and Varadhan2021; Fried et al., Reference Fried, Tangen, Walston, Newman, Hirsch and Gottdiener2001). Physical frailty emerges as a vulnerable state of dysregulation of a complex dynamic system. An important feature is that this is a non-linear process whereby a threshold of aggregate psychological dysfunction is reached, beyond which a state of lower functions of the entire organism occurs.

3.5 Should informal care matter for needs assessment?

In addition to functional impairments, several long-term care systems also consider the amount of informal care (e.g., from partner, adult children or other family members) which is available to the applicant. A general distinction is made between ‘carer-blind’ eligibility rules, where the availability of informal care does not affect eligibility for formal care, and ‘carer-sighted’ rules, where the availability of informal care plays a role (Eleftheriades & Wittenberg, Reference Eleftheriades and Wittenberg2013; Pickard, Reference Pickard2001). Furthermore, carer-sighted approaches can be ‘negative’ or ‘positive’: in a negative carer-sighted system an individual is less likely to be eligible for formal long-term care if informal care is available; in a positive carer-sighted system, the likelihood of being eligible is higher if the claimant receives informal care. The implications and justifications of these approaches are rarely discussed in the literature.

Carer-blind approaches are somewhat more frequent in systems adopting algorithm-based rules with clear-cut rules for eligibility determination, such as in France, Austria, Germany, Belgium and Japan. Conversely, carer-sighted approaches are more frequent in countries where the eligibility rules are more subjective, that is, they are determined under a case-managed approach (Brugiavini et al., Reference Brugiavini, Carrino, Orso and Pasini2017; Fernandez et al., Reference Fernandez, Forder, Knapp, Glied and Smith2011). However, carer-sighted rules are sometimes applied in countries with algorithm-based rules, such as in Italy. Although the Netherlands largely adopts a case-managed system, eligibility for intensive nursing care (the Wlz programme) is ‘carer-blind’, as it depends on the functioning level of the claimant, independent of the informal care they receive.Footnote ¹ However, the Netherlands applies negative carer-sighted rules for some types of home-based formal care: individuals are eligible to receive home-based formal care only when their social network is not capable of arranging sufficient support to fulfil their needs (Alders et al., Reference Alders, Costa-Font, de Klerk and Frank2015). England was traditionally classified as a case-managed system with a minimum safety net and a negative carer-sighted approach. However, since the Care Act 2014, the system is carer-blind (Brugiavini et al., Reference Brugiavini, Carrino, Orso and Pasini2017). Finally, some countries combine elements of a carer-blind and a carer-sighted approach. In France, for example, the availability of informal care does not influence whether someone is eligible for long-term care, but it does affect the amount and content of the care plan. In Italy on the other hand, either negative or positive carer-sighted approaches are employed, depending on the region.

3.6 How eligibility rules affect care coverage

Previous sections in this chapter have shown how the definition of eligibility for public long-term care differs between countries. We now ask a pragmatic question: how do these conceptual differences in needs assessment and eligibility rules affect the potential coverage of long-term care systems? Do different systems target different people, or does legislation mostly differ in language with little effect on the breadth of the different systems? We will show that different definitions of eligibility rules result in very different target populations, that is, the population with a potential access to long-term care support. In section 3.9, we examine how eligibility rules affect long-term care coverage, public budgets and the wellbeing of older people.

Effect of eligibility rules on potential long-term care coverage: the case of European countries

The size and characteristics of the population potentially covered by at least the minimum public long-term care support depends on the definition of ‘need of care’ in the eligibility rules

We will use the term ‘potential coverage rate’ to refer to the share of the population aged 65 and older which fulfils the requirements defined by the rules for accessing public long-term care support. However, due to the complexity of long-term care rules, it is challenging to infer the potential coverage of different long-term care systems through a qualitative comparison of the rules described in section 3.3. Nor would it be useful for our specific aim to compare the rate of actual long-term care utilisation across countries (e.g., the percentage of long-term care users in Germany vs Spain), as countries differ not only in their eligibility rules, but most importantly also in the need for care in their population – a population in poor functional health will report higher rates of long-term care utilisation than a country whose population is in good functional health, even if their systems have the same eligibility rules. To understand potential coverage, therefore, we used individual-level data from the Survey of Health Ageing and Retirement in Europe (SHARE), a large survey conducted among people aged 50 years and over in Europe since 2004 (Börsch-Supan et al., Reference Borsch-Supan, Brandt, Hunkler, Kneip, Korbmacher and Malter2013). SHARE includes detailed self-reported information on sociodemographic, income and health characteristics. We focused on a representative samples of individuals aged 65 and older in eight European countries, namely, Austria, Belgium, the Czech Republic, France, Germany, Italy (national programme), Poland and Spain. While our findings are informative for countries around the globe, we chose these countries because they have well-developed long-term care systems and they apply clear-cut long-term care eligibility rules (Brugiavini et al., Reference Brugiavini, Carrino, Orso and Pasini2017).

We estimated the potential coverage of long-term care rules through the method of direct adjustment standardisation (Brugiavini et al., Reference Brugiavini, Carrino, Orso and Pasini2017; Schokkaert & Van de Voorde, Reference Schokkaert and Van de Voorde2009), which enables us to compare the rate of coverage of long-term care for individuals while holding their demographic and health characteristics constant. Following Brugiavini et al. (Reference Brugiavini, Carrino, Orso and Pasini2017) and Carrino et al. (Reference Carrino, Orso and Pasini2018), we built a binary index of long-term care eligibility for each person using information from both SHARE and the long-term care legislation. For example, an individual in SHARE (regardless of the country they lived in) would be classified as eligible for the long-term care rules of country A if their socio-medical profile fulfilled the requirements to obtain publicly funded long-term care in country A.

A qualitative analysis of the eligibility rules in the selected countries, briefly summarised in Table 3.5 in the appendix, and in more depth in Brugiavini et al. (Reference Brugiavini, Carrino, Orso and Pasini2017), would highlight, for example, how some rules embed a stronger focus on prevention of the loss of autonomy, while others focus more on a cure (to support the individual after a major loss of autonomy occurred). Rules more focused on prevention typically give more weight to limitations in IADLs and/or mobility (e.g., the Austrian and German rules), as opposed to being focused mainly on ADLs limitations (e.g., Italy and Poland), because IADLs and mobility represent more complex tasks, which tend to need help with earlier than ADL tasks. All the laws are carer-blind (as explained in section 3.5, they do not consider whether informal care is already available). However, it is not fully clear whether some rules will lead to higher coverage than others. Our modelling below will help us to evaluate this.

We have computed the potential coverage rates through direct adjustment standardisation for the selected long-term care programmes and show the results in Figure 3.1. This analysis addresses the following question: if we applied the eligibility rules of each country to all the SHARE respondents aged 65 and over living in the eight countries mentioned above, what proportion of them would be covered by long-term care services according to their different eligibility rules? A higher percentage tells us that a system is more comprehensive in its coverage by virtue of its eligibility rules, as we kept health constant by focusing on the sample underlying population. Coverage is expressed as a percentage of the sample aged 65 and over in the eight European countries. Notice that the denominator for these rates is the average European population, which includes many individuals who do not need care. However, the graph enables us to compare potential coverage rates across countries in a hypothetical scenario in which they would have the same level of health.

Figure 3.1. Potential coverage rate for an average of the European population aged 65+

Notes: We estimate the potential coverage embedded in different long-term care rules, defined as the percentage of older Europeans who would be eligible for long-term care benefits according to the definitions of need of various European countries. Each bar therefore does not represent the actual use of care in each specific country, but rather the share of older Europeans who would be eligible for long-term care according to the rules in each specific country. The European population consists of a sample of 19,880 respondents aged 65+, interviewed in the 6^th wave of SHARE (2016) in Austria (AT), Belgium-Flanders (BE-fl), Belgium-Wallonia (BE-wa), Czech Republic (CZ), France (FR), Germany (DE), Italy (IT), Poland (PL) and Spain (ES). The graph reports more than one time point for the eligibility rules in Austria and Germany, to capture recent changes in legislation.

The results in Figure 3.1 show that potential coverage for long-term care is highly heterogeneous between countries. Our estimates suggest that German rules (as reformed in 2017) cover around 20 per cent of the sample and are more inclusive than the rules in any other country, and more than the older German rules. The 2017 German reform lowered the minimum level of vulnerability that would give access to benefits, which explains why the potential coverage rate increased after 2017. The Austrian government has also reformed their eligibility rules in recent years. Unlike Germany, Austrian reforms aimed at tightening access to long-term care benefits by increasing the minimum number of hours of need required to become eligible. The results are visible in our analysis, which shows that the more recent definition of long-term care need in Austria would cover around 12 per cent of older Europeans, compared to 16 per cent under the old rules. The long-term care coverage rates in the European population would also be significantly lower under the rules in Belgium (Wallonia and Flanders), Spain, Italy and Poland.

A possible interpretation, as suggested by Colombo and Mercier (Reference Colombo and Mercier2012), is that countries with a higher rate of potential coverage (e.g., Germany, Austria) have a more pronounced focus on prevention or at least early intervention, while countries with lower potential coverage place a stronger focus on those with already significant limitations (e.g., Italy, Poland, Belgium). However, not all countries with higher potential coverage necessarily focus on prevention. For example, France’s system is primarily focused on acute ADL difficulties and cognitive impairment, while Spain’s long-term care rules assess needs through a more encompassing approach which combines information on difficulties with IADLs, ADLs and cognitive impairment. Yet the French definition of need has a higher coverage than the Spanish one, which shows that the French system compensates its narrower assessment of need with a much looser threshold of minimum dependence; the Spanish system has a broader assessment of need but stricter eligibility criteria, ultimately leading to lower potential coverage.

Effect of eligibility rules on unmet need for people with disabilities

We now show how the differences in the legislative definition of need translate into different coverage rates for older Europeans in poor functional health. We do this by estimating the probability of not being eligible for public support while having a functional health problem, which we will refer to as ‘unmet need’ risk. We considered five different hypothetical functional health profiles, from less to more severe limitations:

1. 1. people with limitations in at least one IADL but no limitation in ADLs or cognition;

2. 2. individuals with at least one limitation in ADLs;

3. 3. individuals with at least 3 ADLs lost; and

4. 4. those with severe cognitive impairment.

In Figure 3.2, we computed the percentage of individuals in each of those groups who would be eligible to receive public long-term care support according to the definitions of needs in Germany (after 2017), France, Italy (national programme) and Poland. Unsurprisingly, the proportion eligible among the groups with only IADL limitations is much lower than that for the group with ADL limitations, illustrating the focus of current programmes on those with severe limitations. Among those with IADL limitations, around 15 per cent would be eligible to receive support in Germany, compared to only around 5 per cent in France or Italy, and none in Poland. This confirms that the German system is more inclusive of people with less severe limitations – potentially prioritising prevention – than the French, Italian or Polish systems. But even in Germany, the risk of unmet needs is very high among those with IADLs.

Figure 3.2. How alternative long-term care rules would cover older Europeans with ill health

Notes: Sample of 19,880 respondents aged 65+, interviewed in the 6^th wave of SHARE (2016) in Austria, Belgium-Flanders, Belgium-Wallonia, Czech Republic, France, Germany, Italy, Poland and Spain.

Unmet needs are also high for those with ADL limitations, despite their higher level of dependency. The German eligibility rules would cover 85 per cent of people with any ADL limitations, whereas the French rules would cover 57 per cent. In contrast, the Italian and Polish systems would cover 21 per cent and 14 per cent respectively, leaving the majority of people with ADLs without coverage. Among individuals with severe activity restrictions (3+ ADLs lost), both the French and the German system would provide almost certain coverage (100 per cent). Conversely, the Italian and Polish system would provide coverage to less than 40 per cent of affected individuals. This again highlights how, except for Germany, these systems are focused on helping only those with very high levels of limitation.

Among individuals with severe cognitive impairment, coverage with the Italian definition reaches 100 per cent, as it does with the French definition. This is because both systems include cognition as a sufficient condition for obtaining eligibility status. The German system rules would cover 80 per cent of people with cognitive limitations, while the Polish system would only cover 10 per cent of this population.

In conclusion, our analysis suggests that eligibility rules have enormous implications for the level of coverage of long-term care systems. By targeting populations with different levels of need, eligibility rules are also a strategy through which countries operationalise a policy focus on early stages of limitations – potentially conferring importance to interventions in the early stages of functional decline, with the intention of preventing further decline or reversing limitations – as opposed to a policy focus on very advanced levels of functional decline, at which point prevention or reversal of functioning may no longer be possible (Colombo & Mercier, Reference Colombo and Mercier2012).

3.7 How eligibility rules affect the intensity of potential coverage

In the previous section we examined how eligibility rules affect whether individuals are covered for public long-term care, but we ignored the fact that the amount of care for which individuals are eligible may also differ across countries. We will now consider the extent to which different systems vary in the intensity or amount of long-term care they provide. To address this question, we capture differences in the intensity of long-term care for which individuals are eligible, using information on the monetary costs of long-term care between countries. This gives us an indication not only of whether individuals receive care, but also how much they receive.

We have computed the share of total costs of long-term care for different severities and care settings that would be covered by public social protection systems in OECD and EU countries, depending on the claimant’s level of need. As income and wealth can play a role in determining the intensity of government support, we have focused on individuals with a median income (among the income of people aged 65 or over) and no net wealth. Because of the level of detail required to perform this analysis, which includes legislation at the local level, it was not possible to include some countries, and only specific regions or cities where rich information was available were included in the analysis. Further details on the methodology are reported in Oliveira Hashiguchi and Llena-Nozal (Reference Hashiguchi and Llena-Nozal2020).

The results are shown in Figure 3.3. Panel A refers to the costs of home care for someone with low, moderate or severe needs. Panel B refers to the costs of home and institutional care for people with severe needs. Government support tends to be greater for older people with more severe needs. However, systems vary widely in their coverage of total long-term care costs, even for individuals with comparable levels of need. This is in line with the heterogeneous picture that emerged from the previous section: most systems cover a greater share of total costs of home-based long-term care for severe needs than for moderate and low needs. On the other hand, older people with lower levels of need in some countries (for example, France, Italian South Tyrol and Latvia) receive more support than those with moderate needs, and those with moderate needs receive more support than those with severe needs. This is due to limits set on the number of hours of care that can be covered through public long-term care benefits and schemes. Indeed, in some countries and subnational areas, there is no coverage for the costs of home care for older people with median income, nor for those with no net wealth and low care needs.

Figure 3.3. Share of total long-term care costs that would be covered by public social protection for care recipients earning a median income for older people and holding no net wealth, by severity and care setting.

Notes: Low, moderate and severe needs correspond to 6.5, 22.5 and 41.25 hours of care per week respectively. See methodological details in Oliveira Hashiguchi and Llena-Nozal (Reference Hashiguchi and Llena-Nozal2020). Country abbreviations: Korea: Republic of Korea; Netherlands: Netherlands.

It is interesting to note that substantial differences in coverage appear even between countries which share similar stages of economic development and similar geographical locations, such as the Netherlands and Germany, or Ireland and England. Moreover, it is interesting to compare this ranking of countries against the widely used classification of welfare states by Esping-Andersen (Reference Esping-Andersen1990), where countries’ welfare systems are classified as liberal, social democratic or conservative-corporatist. We can broadly note that social democratic countries (Northern Europe) show the highest intensity of long-term care coverage, and that France, Germany and Austria (which are typically grouped together in the conservative group) show similar levels of coverage. However, many other countries which are typically grouped together in standard welfare classifications differ largely in this ranking, as shown, for example, by England, Ireland and California (liberal systems), the Netherlands and Belgium (conservative), Spain and Italy (conservative). This result is in line with emerging evidence that public policies in long-term care (and in social care more generally) tend not to follow the policy design of the broader welfare state in many countries (Bertin et al., Reference Bertin, Carrino and Pantalone2021).

Our analysis suggests that the proportion of total costs of long-term care that public systems cover varies both between and within countries and subnational areas, and across levels of care recipient need. Overall, individuals with higher levels of need tend to receive higher coverage from the public long-term care system, yet there are several countries in which intensity does not increase with need of care. Moreover, international differences in long-term care coverage often do not match the general assumptions of welfare states classifications.

Finally, as we shall see in section 3.9, the differences in eligibility rules we just outlined do not only impact coverage for long-term care, but they also have significant implications for the quality of life and wellbeing of older people with limitations and are therefore critical in achieving the goals of long-term care systems.

3.8 Consequences of means testing and eligibility rules

So far we have focused on analysing the impact of eligibility rules on coverage of long-term care. But a separate objective of long-term care systems is to achieve equity, which might potentially be more efficiently achieved by focusing on individuals with lower income or wealth who are unable to purchase long-term care in the market. In addition, while there may be good reasons to invest in more accessible and affordable long-term care, all countries and subnational areas have limited fiscal space. In order to protect older people from the risk of poverty and hardship associated with having long-term care needs, governments often target long-term care support to individuals with limited economic resources. As a result, many countries make eligibility for long-term care dependent not only on functional health needs but also on means testing, targeting government support to applicants with lower incomes and net wealth, while limiting – or even withholding – support from care recipients who can afford to pay more. Means tests can be used to determine both eligibility for long-term care and the level of support for those who are eligible. Different types and combinations of means – income, assets of various types – can be used. Put simply, if a long-term care benefit or scheme is means-tested, then the level of government support and thus the OOP user contributions are dependent on the means of the older person applying to receive care. Means tests can promote equity at the same time that they promote allocative efficiency. In some systems, means-tested benefits and schemes can be combined with non means-tested benefits and schemes (as seen in Table 3.1). However, it is known that this approach can also generate inefficiencies and drawbacks such as limited take-up, poverty traps (caused by high effective marginal tax rates), and stigmatisation (Cremer & Pestieau, Reference Cremer and Pestieau2018).

In England, Germany, South Tyrol (Italy), the Netherlands, Spain and Slovenia, public home care benefits and schemes mix both income and assets testing with non means-tested support. Vienna (Austria), Finland, Reykjavik (Iceland), Latvia and Lithuania combine non means-tested and income-tested only (not assets-tested) benefits and schemes. Ontario (Canada), the Czech Republic, Ireland, Luxembourg and the Slovak Republic have only non means-tested benefits and schemes for home care, while Tallinn (Estonia), Hungary, Japan, the Rep. of Korea and Sweden provide only income-tested (not assets-tested) benefits and schemes. No system combines all four categories of means testing and non-means testing, either for home or institutional care. The most frequent combination in institutional care is both income and assets-tested benefits and schemes, as in Hungary, Ireland, Japan, the Netherlands, Spain, and California and Illinois in the United States. Income testing is the most widely used option, often combined with assets tests, for institutional care in twenty-four countries and subnational areas, and in-home care in twenty-one countries and subnational areas.

The case of assets testing is particularly relevant from a life-course economics perspective. As with other types of means tests, the underlying assumption is that older people with higher net wealth are better able to afford the OOP costs of care; even if they might not be able to afford costs from their incomes alone, they are likely to be able to cover the shortfall from their assets. However, in order to qualify for greater government support, older people with care needs in countries and subnational areas with means tests may need or decide to deplete their assets to reach a threshold that qualifies them for increased public support (a process sometimes referred to as ‘strategic impoverishment’, see Cremer & Pestieau, Reference Cremer and Pestieau2018).

Assets-tested long-term care benefits and schemes are common in both home care and institutional care in OECD and EU countries and subnational areas. Assets tests tend to share certain features (see Table 3.2). Certain countries and subnational areas set wealth thresholds for greater government support, but these tend to be low compared to national mean net wealth among older people. Above these thresholds, some benefits and schemes take into account only a share of total assets when determining the level of government support. Different types of assets may be treated differently depending on the country, subnational area and even specific long-term care benefit or scheme. The primary residence is frequently excluded from assets tests, especially in the case of home care as it is expected that the person receiving care is still living in their home. Deferred payment agreements may be used, in which case older people receiving home care agree to use their assets (including their primary residence) to repay the public social protection system for OOP costs incurred while living in their home if they sell their house (e.g., when moving to institutional care) or when they die.

Assets testing might however incentivise distortionary behaviours in terms of saving decisions that people take. We discuss this in Box 3.1.

3.9 Do differences in eligibility rules affect care utilisation and wellbeing of older people?

Eligibility for long-term care increases the probability of receiving care among older people

Figure 3.4 shows that, as we would expect, the probability of receiving home-based care increases as the number of functional limitations with ADLs increases. Most importantly, it shows that the probability of receiving home-based care is higher for individuals who are eligible for public long-term care support (the red bar) than for individuals who are not eligible (the blue bar). It is very important to stress that this is true even for individuals who have the same number of ADL limitations, e.g., among those with only one ADL limitation, the probability of receiving home-based care is just below 50 per cent for the eligible group, compared to less than 20 per cent for the non-eligible group.

Figure 3.4. Proportion of respondents receiving formal home care, by number of ADL/IADL limitations and by long-term care eligibility status

Note: Sample consists of 19,880 respondents aged 65+, interviewed in the 6^th wave of SHARE (2016) in Austria (AT), Belgium-Flanders (BE-fl), Belgium-Wallonia (BE-wa), Czech Republic (CZ), France (FR), Germany (DE), Italy (IT), Poland (PL) and Spain (ES). The graph reports more than one time point for the eligibility rules in Austria and Germany, to capture recent changes in legislation.

We performed a full statistical exercise to estimate the effect of being eligible on the probability of receiving home-based long-term care, where we controlled for such confounding variables as socio-demographics, health and functioning status. Our results suggest that being eligible for public long-term care support increases the probability of receiving home-based care by twelve percentage points (baseline level, 7 per cent): for an individual of average health, the probability of accessing long-term care is 19.7% if they are eligible, compared to 7.4% if they are not. This implies that a change to eligibility rules can substantially increase – or decrease – the likelihood that people will be covered for long-term care. This is particularly relevant for policy makers, as it shows that eligibility rules play an important role in reducing the risk of unmet needs among older people, above and beyond individual characteristics such as education, income wealth and health.

Are some groups more likely to gain access due to their eligibility status?

Eligibility for long-term care seems to act as a strong incentive to access formal long-term care. However, there might be some groups of individuals for whom the incentive is stronger. We are particularly concerned about individuals who might fail to access formal care despite being eligible for it (and therefore face a higher risk of unmet needs). Our analysis identified two groups at higher risk of unmet needs, who would particularly benefit from policy support:

1. 1. Our evidence suggests that people with lower levels of education are less likely to access formal care even after they become eligible for it, and hence face a higher risk of unmet needs. In Figure 3.5 (first panel) we show that, comparing people with similar health status, those with lower levels of education (up to lower secondary degree) have roughly the same probability of using care as higher educated groups, if they are not eligible for long-term care programmes. While all groups are more likely to use care if they are eligible for public long-term care, the increase in care use induced by eligibility status is significantly higher for those with higher education. This may suggest that people with higher education have a stronger preference for the use of formal care than their less educated counterparts. However, the fact that people with lower levels of education are less likely to take up formal care even when they become eligible for it, compared to people of higher education and with similar health, suggests that policy makers should be very conscious of differences in long-term care literacy, economic endowments, and individuals’ ability to navigate the complex mechanisms or rules and points of access (García-Gómez et al, Reference García-Gómez, Hernández-Quevedo, Jiménez-Rubio and Oliva-Moreno2015; Rodrigues et al., Reference Rodrigues, Ilinca and Schmidt2017).

2. 2. Our evidence also suggests that, among older people eligible for formal home-based care, those living alone are more likely to receive it than people living with their relatives. The impact of long-term care eligibility on care utilisation varies by parenthood status: compared to non-eligible respondents, people eligible for long-term care but with no children increase their likelihood of accessing care much more than people with children (Figure 3.5, third panel). This suggests that the availability of a family network, which is the main source of long-term care support worldwide, might constitute a partial disincentive to seek formal support, due to the preference of many older people to be cared for by family members rather than external professionals (Kalwij et al., Reference Kalwij, Pasini and Wu2014). While this could reflect the preference of older people, it might also highlight that such individuals are missing out on potential help from the state. There are at least two reasons why governments should be concerned by this result. First, the absence of formal support may deprive older people of access to higher-skilled care, often provided by formally trained caregivers in comparison to informal ones. Second, it highlights the risk of increasing the psychological and physical burden on caregivers (including work/family conflicts), who could have been partially relieved or helped by publicly subsidised care.

Finally, we show that eligibility for long-term care increases care use in a similar way for both sexes and for people with different levels of wealth (Figure 3.5, second and fourth panels).

Figure 3.5. Differential impact of eligibility for long-term care on care use, by sociodemographic characteristics

Note: Sample consists of 19,880 respondents aged 65+, interviewed in the 6^th wave of SHARE (2016) in Austria (AT), Belgium-Flanders (BE-fl), Belgium-Wallonia (BE-wa), Czech Republic (CZ), France (FR), Germany (DE), Italy (IT), Poland (PL) and Spain (ES). The graph reports more than one time point for the eligibility rules in Austria and Germany, to capture recent changes in legislation.

3.10 Policy experiment: how would care utilisation among older populations change if governments adopted different eligibility rules?

How can policy changes in eligibility rules affect the demand for domiciliary long-term care? In Figure 3.6 we show how the current percentage of long-term care users in eight European countries would change if governments reformed their long-term care eligibility rules for home care support by shifting the focus from providing assistance for severely impaired people to prevention of loss of autonomy. We evaluated the impact on care utilisation of two scenarios in particular: (i) governments reform eligibility rules so that anyone with an ADL limitation is eligible; or (ii) governments reform eligibility rules so that anyone with limitations in either ADL or IADL tasks is eligible. These scenarios are compared to the current status quo, i.e. the share of older adults currently using formal home care under the current rules. Results show that, in general, such changes would significantly expand the population of formal home care users. The effects vary between countries, due to the different prevalence of functional limitations and the different extent to which current eligibility rules already cover ADL and IADL limitations. Allowing the eligibility rules to cover all ADL limitations would lead to an increase in actual access to care by around 2 percentage points in Italy, Spain, France, Belgium and Poland, whereas extending the eligibility rules to all IADL limitations (i.e. an approach more focused on prevention) would increase access to care by a minimum of 2.5 points and a maximum of 4.6 points.

Figure 3.6. Eligibility rules more focused on prevention (IADL limitations) compared to current rules would expand population of care users

Note: Sample consists of 19,880 respondents aged 65+, interviewed in the 6^th wave of SHARE (2016) in Austria (AT), Belgium-Flanders (BE-fl), Belgium-Wallonia (BE-wa), Czech Republic (CZ), France (FR), Germany (DE), Italy (IT), Poland (PL) and Spain (ES). The graph reports more than one time point for the eligibility rules in Austria and Germany, to capture recent changes in legislation.

3.11 Conclusions

All governments have limited fiscal space to provide long-term care support and services, and needs assessments are one mechanism to make sure those limited resources are used in ways that maximise access, affordability, equity (i.e. the most vulnerable are adequately protected) and efficiency (i.e. a given level of results is achieved at the lowest cost to the public purse). This chapter has focused on the lessons learned from OECD countries with advanced long-term care systems, many of which are European. Even though all countries explored in our analysis offer long-term care services and support, these programmes do not achieve comprehensive coverage due to the nature of their eligibility rules.

This chapter has explored the economic and wellbeing consequences of long-term care eligibility rules, with a particular focus on home-based care. Eligibility rules of public care can substantially affect access to home-based care among older people. Eligibility to receive care can also offer social protection from the financial risks associated with functional decline and the need for long-term care, contributing to reducing the risk of poverty. Long-term care coverage through public eligibility can also improve the mental wellbeing and life satisfaction of older people, thus delivering better outcomes. We have shown that, in most countries with mature long-term care systems, eligibility for home-based care targets advanced levels of disability, with a focus on compensation or rehabilitation rather than prevention. Going beyond disability and incorporating measures of frailty may enable governments to intervene early and prevent a spiral of decline, which may improve long-term wellbeing outcomes and reduce costs of long-term care. Moreover, eligibility rules are sometimes complemented with means assessments, which prioritises groups with lower economic resources.

We outline here four major takeaways from our study.