Chapter 4. Health and long-term care services for advanced dementia are poor

This chapter examines what countries are doing to provide better care for people with advanced forms of dementia, including in care facilities, hospitals, and at the end of life. It finds that the quality of care for people with advanced dementia is in many cases poor. Better measurement, including strengthened hospital coding and more regular monitoring of quality of care measures, is important to ensure that best practice policies are more routinely implemented, particularly in formal care settings.


The statistical data for Israel are supplied by and under the responsibility of the relevant Israeli authorities. The use of such data by the OECD is without prejudice to the status of the Golan Heights, East Jerusalem and Israeli settlements in the West Bank under the terms of international law.

Though many countries have increasingly focused on community-based care and ageing in place, people with dementia nevertheless require significant professional support. People living at home with moderate dementia may need at least occasional help with basic activities of daily living, such as washing, dressing, and preparing meals, while many with more advanced forms of the disease will also require more intensive care support, including full-time home health aides or nursing staff. Those living in nursing homes or other long-term facilities will rely even more intensively on professional care support. People with dementia, and particularly advanced dementia, often exhibit psychological and behavioural symptoms from the disease and may become agitated or otherwise difficult to manage, particularly when confronted with new situations or people. Caring for people with dementia, who in addition to behavioural challenges may be unable to communicate their desires or needs, can be particularly difficult. Basic training programmes for professional health and social care staff may not adequately prepare professional carers for the challenges of caring well for someone with dementia. Dementia-specific approaches that take a person-centred approach to caring are critical for ensuring people with dementia require high-quality care from professional care workers and in long-term care institutions.

4.1. Ensuring better facility-based care for people with dementia

Traditional long-term care settings are often ill-suited to caring for the needs of people with dementia, and countries should consider how existing facilities can be adapted and new facilities can be developed to better meet the needs of people living with dementia. People living with dementia often have unique care needs that can be ill-suited to the traditional model of institutional care. In recent years, a better understanding of their needs has led to a re-evaluation of traditional models of long-term institutional care. Emerging models, including small-scale living communities and ‘dementia villages’ provide a more community-based model of support, even within institutional care settings.

However, driving further improvements in the quality of care offered in long-term care facilities is hampered by persisting data constraints. Dementia remains grossly underdiagnosed among people living in care facilities, and almost no countries have good data on the prevalence of dementia in care facilities. Nevertheless, estimates suggest that a majority of nursing home residents have cognitive impairment and dementia. Studies in Europe (Sweden, Czech Republic, England, Finland, France, Germany, Israel, Italy, the Netherlands), for example, indicate that close to 70% of nursing home residents have some form of cognitive impairment (Onder et al., 2012; Björk et al., 2016). Yet just 10 of 21 OECD countries reported that they can use administrative data to identify people with dementia in long-term care facilities. Without the ability to identify people with dementia, it will be difficult for countries to measure outcomes or quality of care and to determine whether policies put in place to improve care are working.

4.1.1. Institutional care must adapt further to meet the needs of people with dementia

New models of institutional care are emerging that are specialised to meet the needs of people with dementia

Small-scale living communities offer an alternative to traditional nursing-home facilities and have been adopted across the OECD. These arrangements usually comprise of living environments where people with dementia share a home with other residents and staff. Such arrangements are intended to facilitate communal living and social engagement more than traditional nursing home arrangements, with a view to improving residents’ quality of life and providing them with more independence than in traditional facilities. This person-centred approach expands the focus beyond basic clinical care to facilitating the social and emotional needs of people living with dementia.

Across the OECD, about 4% of people 65 and over – and more than one in ten people over 80 – live in care facilities (see Figure 4.1). While most surveyed OECD countries reported having at least some small-scale living facilities available in their country, no countries have transitioned the majority of their residential care services to this model. Most people, including those with dementia, continue to reside in more traditional facilities. Spaces in innovative facilities, including small-scale living communities, multi-generational homes, and dementia villages are limited. Moreover dementia-specialised long-term care facilities providing the latest models of care are often private. The high costs of private nursing home care frequently creates a significant barrier to access, particularly when no- or low-cost facilities are also available.

Figure 4.1. Long-term care recipients in institutions, 65 and over
% of population aged 65 and over, 2015 (or nearest year)

Source: OECD Health Statistics 2017,


Figure 4.2. Long-term care recipients in institutions, 80 and over
% of population aged 80 and over, 2015 (or nearest year)

Source: OECD Health Statistics 2017,


Countries should consider developing or adapting existing guidelines on dementia-friendly design to be used for new construction and renovations. The physical design of buildings can strongly impact the quality of life of people living with dementia. In more ‘traditional’ residential facilities, long hallways, nursing stations, and other aspects of traditional nursing home design can contribute to increased anxiety and agitation, while locked doors and other measures taken to prevent wandering can restrict autonomy and create distress. As attention to dementia has grown, a number of countries have developed design principles to incorporate into institutional care and other settings frequented by people living with dementia. Some OECD countries, including the Denmark, Ireland, Norway, and the United Kingdom (England), have developed design guidelines and advice specifically related to dementia.

Box 4.1. Improving care for people with dementia in care facilities

In 2013 and 2014, the Department of Health in England supported 42 NHS and 74 Social Care pilot projects with GBP 50 million with the goal of developing models that improve care for people with dementia. Funding was provided to range of facilities, including care facilities, hospitals, day and respite centres, general practice offices, and dementia care facilities. 73% of NHS and local authority pilot projects took place in care homes. Based on the results of the pilot studies, Loughborough University, with support from the Department of Health, developed a set of best practice guidelines on designing dementia-friendly environments. This guidance was published in 2015 as a Department of Health building note for health and social care environments.

In some cases, financing is available for or even tied to compliance with dementia-friendly design. Norway, for example, requires that residential and nursing homes built with funding from the National Bank of Housing conform to dementia-friendly design principles, while grants and subsidies are available for renovating existing homes.

More research is needed to build the evidence base on alternative models of long-term care

While the development of small-scale living facilities shows promise, more evidence on the effect of these new living arrangements on outcomes is needed. The limited research that has been conducted so far has demonstrated inconclusive though promising results in improving residents’ quality of life. Studies of the ‘Green House’ model of small-scale group living for people with dementia in the United States have indicated that quality of life among residents of Green House communities was higher than that of residents in more traditional nursing homes, and quality of care was as good as or better than in traditional nursing homes (Kane et al., 2007). Small-scale group living facilities have been found to use far fewer restraints and psychotropic medications, a further suggestion that quality of care may be higher (Verbeek et al., 2014). Family members have also reported higher satisfaction and lower burden in small-scale group homes than traditional facilities (Lum et al., 2009, Verbeek et al., 2010). However, other evaluations have suggested no differences in the quality of life of residents in small-scale or traditional nursing facilities (Verbeek et al., 2010, Auer et al., 2017). Though the current evidence is mixed, the evidence related to higher quality of life and care is encouraging. Countries must continue to develop alternatives to traditional long-term care institutions that expand the notion of care beyond clinical improvement, to include promoting the autonomy of people living with dementia and improving their quality of life to the greatest extent possible.

4.1.2. More dementia-specific training for care staff could lead to better quality care

Substantial dementia training is rarely included in the minimum training requirements for care staff

Dementia-specific training is associated with improvements observed in how staff reacted to problems with long-term care residents (Teri et al., 2005; Spector et al., 2013). A study of Australian health care staff suggested that there were gaps in staff knowledge about dementia, particularly where they had not received specific training or cared for people with dementia (Smyth et al., 2013). Studies of training programmes delivered to formal care workers and staff members at nursing homes and other long-term care institutions indicate that dementia-specific training programmes can have a positive impact on both the staff’s ability to handle difficult behaviour, and the prevalence of behavioural challenges in the first place (Livingston et al., 2005). The frequency of behavioural symptoms related to dementia have been associated with the level of staff training, among other factors, and interventions delivering staff trainings have been found to have significant impacts on reducing agitation and other behavioural challenges (Boustani et al., 2005; Chenoweth et al., 2009; Spector et al., 2013Even where professional care staff are fairly knowledgeable about dementia, there is some evidence to suggest that confidence in handling difficult situations can be low, and can be raised through dementia-specific training (Hughes et al., 2008). The level of training and experience of staff members may also impact their attitude toward people with dementia. Staff members with more training are more likely to have a positive outlook despite patient aggression, suggesting that education is helpful in changing attitudes and possibly behaviour toward people with dementia, particularly those with behavioural problems (Nakahira et al., 2009).

Formal care workers in most OECD countries are required to undergo a minimum level of training which does not always include specific information related to caring for people living with dementia. Though many OECD countries offer voluntary dementia-specific training programmes tailored for or including professional care staff, few countries have dementia-related requirements for training. Many people in care facilities have or will develop dementia, and may display behaviours that are challenging or risky to themselves and the people who care for them. Knowing how to manage these behaviours is critical to providing high quality and safe care for both people with dementia and their carers. Countries should consider how they can further incorporate dementia care into training programmes for health and social care staff, from low-skilled carers to directors of long-term care institutions. For countries that already have developed training programmes, courses can be evaluated for how well they teach components of person-centred dementia care, and additional modules considered where gaps are identified. In Australia, the move toward increasing privatisation of aged care services has incentivised aged care providers to differentiate themselves and distinguish themselves based on both available services and quality of care, including by enrolling their staff in dementia training programmes offered through Dementia Training Australia, a national government initiative implemented in partnership with universities and Dementia Australia (Box 4.2).

In some cases, dementia training is covered at least partially by basic training programmes, as in the case of training requirements to achieve different levels of qualification through the National Certificate in Health, Disability and Support in New Zealand. In Belgium, basic dementia training is required to work as a dementia-aware home-care worker. Home care organisations in Belgium must have at least some professional home workers who have received training that prepares them to provide basic home care assistance to people living with dementia. About 21 hours of in-person training is required for the position, which otherwise has a relatively low educational requirement. Since the programme was launched two years ago, more than 1 200 people have undergone training to become dementia-aware home-care workers. In Germany, a new professional care position – that of an additional care helper was established in 2009 to provide additional support in nursing homes. The position was developed out of an understanding that greater social interaction and support is important for the well-being and quality of life of people living with dementia, including those in nursing homes and other long-term care facilities. Additional care helpers are required to undergo a short training course of approximately six weeks to qualify for the position. They are not allowed to participate in direct care needs, including washing, dressing, and nursing. To incentivise participation in this new position, additional care helpers receive a special minimum wage that is higher than the legal minimum wage of each Bundeslände. Though originally developed for people with dementia, the programme was expanded in 2016 to include all people living in care homes, with a requirement that one additional care helper be available for every twenty residents.

In the United Kingdom (England), all staff members who work in organisations interacting with people with dementia are required to undergo dementia training in line with their level. The Dementia Core Skills Education and Training Framework was developed to detail the level of knowledge about dementia expected of different staff members working with people with dementia. Learning outcomes are detailed for all levels of staff, and comprise three tiers: baseline dementia awareness training (all staff); training for staff members who interact regularly with people with dementia; and training for service leaders and other experts. In addition, all health and social care staff members are required to complete at least Tier 1 dementia awareness training. Approximately 85% of social care staff have completed a care certificate for dementia, with 55% of the workforce completing a qualification of at least Tier 2.

Box 4.2. Developing national dementia training programmes in Australia

Australia offers two nation-wide training programmes for caring for people with dementia. The first, Dementia Training Australia, has made available professional dementia care development training, opportunities for upskilling, and access to accredited dementia care training courses across the country. Accredited dementia vocational courses are available at no cost to eligible care workers, including in both institutional and community-based care. Dementia Training Australia offers training opportunities across health and social professions, including courses aimed toward general practitioners, nurses and pharmacists in addition to care workers. Online training is also available. In addition, aged care providers who are interested in improving the quality of their dementia care are able to request onsite training interventions. Following an audit of the care home environment and current dementia skills, a targeted training package aimed is developed to address weaknesses identified during the audit.

The programme is run by the Australian government together with five universities and Dementia Australia. Building on programmes previously available in different parts of the country, the national platform was launched in 2016. Dementia Training Australia offers an online resource portal for health and social professionals and aged care service providers to identify available opportunities for training and accreditation.

In addition to the training opportunities offered through Dementia Training Australia, the Dementia Behaviour Management Advisory Service (DBMAS) offers assistance to both individuals and service providers when behavioural and psychological symptoms of dementia (BPSD) affects quality of life or their ability to care. DBMAS is a comprehensive, free service that offers support through a number of approaches, including individual assessment, care planning and short-term case management, mentoring, and capacity building and training for service providers.

In Denmark, nursing homes are eligible to apply for public funding to finance dementia-specific training for their staff. DKK 145 million (about EUR 19.5 million) have been allocated to support training for dementia care aides. Dementia training is similarly state-financed in Norway, where municipalities – whose memory teams offer assessment services and pre- and post-diagnostic support to people with dementia living in the community – can apply for grants to send their staff members to dementia training. The ABC dementia training programme in Norway has been found to increase knowledge about dementia and job satisfaction. Over 19 000 basic care staff have completed the programme in Norway, and similar programmes have since been adopted in other countries, including Denmark and Sweden (Box 4.3). In the Netherlands, home and institutional care workers are required to undergo basic training. The uptake of dementia-specific training programmes is largely dependent on care providers.

Box 4.3. The Dementia ABC programme in Norway

The Dementia ABC educational programme developed in Norway represents an example of how well-designed training programmes can be effectively integrated into existing health and social services to deliver high-quality care and improve job satisfaction. Developed by the Norwegian National Advisory Unit on Ageing and Health, the Dementia ABC programme is a two-year, interdisciplinary training programme open to professional staff working with people with dementia, across disciplines and levels of seniority. It is a national programme, run by the Ageing and Health unit, and works with a network of regional centres focused on dementia care education, facilitating accessibility across the country. Participants first receive informational booklets, developed by dementia care experts, which cover components of evidence-based dementia care, including engaging with both the person with dementia and their family members. Local discussion groups are organised to meet regularly to discuss the subjects discussed in the booklets and explore how these themes relate to the work of participants. Lastly, annual workshops are organised by the regional dementia care education networks. The programme was designed around a highly person-centred model of care, and emphasises the central role of the person with dementia and their families.

To promote engagement with the Dementia ABC programme, the Norwegian federal government, through the Norwegian national dementia plan, has engaged in a cost-sharing scheme with local municipalities. Municipalities pay for the informational dementia booklets and are responsible for financing the group meetings and time staff spend attending group Dementia ABC meetings. Participants in the ABC programme are compensated by their employers for the time spent attending the group meetings in about half of cases, while others attend meetings outside of their work hours. Annual workshops, which are offered twice per year, are financed through the national dementia plan.

The programme is not targeted at a specific group of professionals, but is designed to be relevant for all who may engage with people with dementia in their work lives. A longitudinal study of the Dementia ABC programme was recently undertaken using data from 90 municipalities across Norway. Of the nearly 1 800 participants included in a longitudinal survey of the programme, 26% had a bachelor’s degree, 61% were auxiliary nurses with three years of high school nurse training, and 10% had no formal health or social education – representing housekeepers, gardeners, nursing assistants, and other professions. Nearly 80% of participants came from long-term care institutions, including nursing homes and special care units, with 20% providing home-based care and 2% working in day care facilities. Notably, nearly all participants – 97% – were women.

Staff members who completed the educational programme were found to have statistically significant increases in practicing person-centred care and exhibited improvements in job satisfaction. The positive effects of the programme persisted even after the educational programme was completed. Similar programmes have since been adopted in Sweden and Denmark.

Source: Rokstad et al (2016); Kristiansen (2016).

Financial incentives and certification can help encourage participation in dementia training, particularly among more junior care staff

In most cases, even where training programmes are available, health and social care professionals working with people living with dementia may lack all but personal incentives to participate. Professional caring is largely a high-stress, low-skilled and low-paid profession. Care workers may be wary of enrolling in additional training programmes at their own expense – both in terms of time and money invested – where the professional benefit of doing so is unclear. Most countries have not developed incentives for care workers to participate in skills training for dementia.

Developing recognised accreditation programmes that provide health and social care workers and service providers with a clear professional advantage can be considered by countries interested in improving their quality of dementia care. In Sweden, the Swedish Dementia Centre has launched a dementia certification programme for entire care units. Nursing homes, day care centres, hospitals, and other facilities are eligible for certification, which is achieved through an integrated training programme that focuses both on training staff in dementia competence and building strong leadership in the institution. Since the programme was launched, 56 home care units in Sweden have been certified, as has a day care centre in Germany and nursing home in Poland.

Some academic institutions have begun offering official classes and degrees in dementia care, helping to further formalise the discipline. In Australia, the University of Tasmania recently launched a series of dementia care degree programmes. Students can receive a diploma, associate’s degree, or bachelor’s degree through the coursework, which is conducted entirely online and can be undertaken both part- and full-time. The programme is open to people from a variety of backgrounds, including health professionals and carers. The University of Tasmania has also developed a massive open online course (MOOC) on understanding dementia. Since 2013, more than 120 000 people have enrolled in the online course, with some enrolling in dementia care degree programmes following completion.

Universities in Sweden have similarly brought dementia care into formal education with the development of a series of degree programmes. Advanced degrees in dementia care have been developed for doctors, nurses, as well as physiotherapists and occupational therapists. Dementia care study programmes have been designed as part-time programmes, to allow health professionals to continue working while undertaking advanced study. In Sweden, universities including the Karolinska Institute and Sophiahemmet University have increased the reputation of the master’s degree programmes by partnering with the Stiftelsen Silviahemmet, a dementia care foundation headed by Queen Silvia of Sweden. Graduates of the programmes receive their degree and receive the title of Silvia Doctor, Silvia Occupational Therapist, Silvia Physiotherapist, or Silvia Nurse. Since the programmes were started, more than 600 nurses have completed the master’s degree.

Professional care workers are typically very low paid. Few carers receive substantially more than the federally or locally-mandated minimum wage, and many work long hours, often working overtime hours with no or limited additional pay. In the United States, for example, the wages of professional care staff have been largely stagnant over the past decade, despite growing demand (PHI, 2017).

Figure 4.3. Median hourly wage by long-term care setting (United States)

Source: PHI (2017).


Low compensation, coupled with the high stress and low social prestige of caring, often creates low job satisfaction and high turnover among carers. With the need for caring increasing with the growing elderly population, governments face opposing interests when considering how to address the issue. Raising salaries may increase the attractiveness of caring as a profession, but will also increase the costs of health and social care for governments during a time at which many are already struggling with stagnant or falling budgets.

Few countries have established salary bands for professional carers independently of applicable minimum wages. An important exception is New Zealand, where the 2017 Care and Support Workers Pay Equity Settlement Agreement has recently led to a substantial increase in the salaries of professional carers. The settlement, which resulted from a gender-discrimination claim brought by a female carer in New Zealand, has established a pay structure that sets a minimum hourly wage based on the experience of the care professional. Notably, the salary thresholds are determined by both the number of years of caring experience and the qualification levels of the carer. Carers who have achieved level 4 (senior support) of the National Certificate in Health, Disability, and Aged Support, for example, will receive a minimum salary equivalent to carers with no qualifications but 12 or more years of practical experience. These financial incentives may impact dementia care: Level 4 in the National Certificate in Health, Disability, and Aged Support receive advanced training on dementia care beyond what is taught in earlier levels of training.

Salary incentives are also in place for caregivers in Norway. Home health aides and other professional care staff will be eligible for higher salaries contingent on completion of vocational training courses focused on improving dementia care. Special minimum wage rules are also in place for additional care staff in Germany.

4.1.3. The appropriate management of behavioural and psychological symptoms of dementia (BPSD) remains a major challenge for dementia care

As the severity of dementia advances, many people living with the disease begin to experience behavioural and psychological problems, which can manifest as aggressive or agitated verbal and physical behaviour, and sometimes psychosis. Aggressive behaviours and other symptoms of BPSD have been treated by the use of physical or pharmacological restraints. This approach may prevent a person with dementia from harming themselves or others, but seriously infringes on their well-being and independence. They can also have negative health consequences: antipsychotic use can increase the risk of stroke, falls, pneumonia, and hasten cognitive deterioration (OECD, 2015a).

Despite the efforts of policymakers, the use of antipsychotics remains common and prescribing rates vary by a factor of two across OECD countries

Antipsychotic medications, which supress the behaviours by tranquilising the patient, have been frequently used to manage difficult behaviour among people living with dementia. However, many countries recommend against the practice. Antipsychotics are associated with higher mortality and a higher risk of developing strokes and other cerebrovascular events. NICE guidelines in the United Kingdom advise against administering antipsychotics to people with dementia unless they are under severe distress, or pose a risk to themselves or others. Moreover, there are ethical concerns associated with sedating people with dementia when non-pharmacological treatment options are available. Nonpharmacological interventions have furthermore been found to be effective for addressing BPSD in people with dementia (Ayalon et al., 2006). Given the widespread agreement that antipsychotic medications should not be administered to manage BPSD, rates of antipsychotic prescribing can be considered a useful indication of the quality of care people with advanced dementia are receiving. Because of difficulties in estimating the number of people living with dementia, establishing an accurate cross-national estimate of antipsychotic prescribing rates among the dementia population is challenging. Nevertheless, evidence from Swedish registry data suggests that the majority of antipsychotic prescribing for people 65 and older goes to people with dementia, suggesting that the rate of antipsychotic prescribing in older populations can be considered a useful proxy measure.

Figure 4.4. Antipsychotic prescribing rates across OECD countries
Per 1 000 people aged 65 and over

1. Data for Norway do not include people in institutional care, so underestimates the use of antipsychotics. 2. Data for Spain refer to 2014.

Source: OECD Health Statistics 2017,


Across the OECD, prescribing rates for antipsychotic medications in the population aged 65 and over varies by a factor of more than two, from 30 in every 1 000 people aged 65 and over in Sweden and the Netherlands to more than 70 in Belgium and Slovenia (Figure 4.4). Encouragingly, most countries have seen prescribing rates decline over time. The majority of countries have seen stable or decreasing rates, with Denmark showing a particularly marked fall over the past eight years. While Belgium has one of the highest rates of prescription of antipsychotics, the last two years have seen a reduction from 73 per 100 people to 70. However, Slovenia shows a different pattern to other countries, with the use of antipsychotics rising steadily from 63 per 1 000 persons 65 and above in 2009 to 75 per 1 000 in 2015 – a 20% increase. This could raise concerns about the overuse of these drugs to manage difficult behaviour in older people.

Antipsychotic prescribing rates also show a marked difference between men and women. In every country surveyed, women were prescribed antipsychotics more frequently than men. In Slovenia, the prescription rate among women 65 and over was more than 50% higher than among men in the same age group (Figure 4.4). Differences in prescribing rates by sex do not appear to be driven by the age structure of women and men. In 14 of the 15 countries that reported data, women had a higher rate of antipsychotic prescribing in every five-year age group 65 and above. The only exception was the Netherlands, where women aged 80 and over had a lower prescription rate for antipsychotics than men in the same age group.

Figure 4.5. Antipsychotic prescribing rates, by sex
Per 1 000 people aged 65 and over, 2015 (or nearest year)

1. Data for Norway do not include people in institutional care, so underestimates the use of antipsychotics. 2. Data for Spain refer to 2014.

Source: OECD Health Statistics 2017,


Beyond the work on antipsychotics, there is limited internationally-comparable data on quality indicators in long-term care. One recent effort to compare the prevalence of quality of care measures in nursing homes and hospitals has used point prevalence studies to compare the prevalence of different measures of quality in long-term care facilities, including the use of restraints. Results from the Netherlands and Austria suggest that the prevalence of physical restraints use in nursing homes remains high, at over 25% in both countries in 2012 (Halfens et al., 2013). Taken together, the recent findings on antipsychotic usage and the use of physical restraints in the Netherlands and Austria indicate that the quality of care in nursing homes remains suboptimal for many people, including those with dementia.

Some countries have developed guidelines or regulations around the use of restraints and antipsychotics

A number of OECD countries have explicitly banned or regulated the use of restraints in institutional care settings. Financial incentives may help to encourage compliance with regulations. In the United States, funding for nursing homes and other long-term care institutions from the Centers for Medicare and Medicaid Services is tied to adhering to rules related to restraints and locked wards.

Training of health and social care professional staff can also reduce the use of restraints. Staff who have been trained in alternatives to managing behavioural and psychological symptoms of dementia have been found to significantly reduce their use of restraints in institutional settings (Testad et al., 2005; Köpke et al., 2012). Less experienced staff members have also been found to have a more positive view on restraints (Hamers et al., 2008). These findings together suggest that developing clear guidelines and training programmes to teach staff alternatives to applying physical or administering pharmacological restraints can have a positive impact on the quality of life of people living with dementia. Moreover, switching away from a restraint-based approach to managing BPSD has not been found to increase suboptimal outcomes, such as increases in falls or higher rates of antipsychotic prescribing (Gulpers et al., 2011; Koczy et al., 2011).

Better monitoring of how difficult dementia symptoms are managed could help to reduce inappropriate care

Some countries have begun to monitor measures of care to inform practice and improve quality of care. In addition to the SveDem dementia registry, Sweden maintains a registry for behavioural and psychological symptoms of dementia (BPSD). Launched in 2010, the registry aims to improve the quality of care of people with BPSD in nursing homes (BPSD, 2017). All persons with BPSD included in the register have demographic information, diagnostic information, and BPSD-related care information – particularly the use of pharmacological and non-pharmacological treatments – recorded, with follow-up data recorded at multiple points. The registry facilitates continuous learning and care improvement for care professionals, including providing advice to staff on improving care based on the information entered into the registry (Sweden National Quality Registers, 2016). The registry has contributed to declines in inappropriate use of antipsychotics to manage BPSD, an important component of quality dementia care. Nearly 28 000 individuals have been included in the registry since its inception (Sweden National Quality Registers, 2016). More than half of qualifying nursing homes, care units, and BPSD teams participate in the registry (Emilsson et al., 2015).

4.2. Improving the quality of hospital care

4.2.1. Hospitals in OECD countries are not well adapted to the needs of people with dementia

Dementia is a fatal condition, with a trajectory that progresses even when care is good and all available treatments are administered. This means that even under ideal circumstances, there will be people with dementia who will need to be admitted to acute hospitals for care. Even with excellent community-based care and programmes to reduce or delay hospital admissions, people with dementia will visit hospitals, and the relevant services should be prepared to treat them when they do.

Hospitals are not always well equipped to provide people with dementia with high-quality care. Being admitted to hospital can be a distressing experience under the best of circumstances. For people with dementia, and particularly those with advanced dementia, a visit to the hospital can worsen health outcomes, and they may leave the hospital doing more poorly than when they arrived. For many hospitals, the high proportion of hospitalised patients who have dementia means that improving care for dementia patients is critical to the hospital system as a whole. In England, for example, a quarter of the hospital population is estimated to have dementia.

Outcomes for people with dementia who go into hospital are often poor

Staying in hospital can hasten cognitive and functional decline: people with dementia admitted to hospital have been found to have higher long- and short-term mortality rates than similar populations admitted to hospital who do not have dementia (Sampson et al., 2009; Morrison and Siu, 2000; Nightingale et al., 2001). These differences in outcomes between people with dementia and similar populations do not appear to be fully explained by the physical state of people with dementia. Rather, both the hospital environment and differences in the quality of care received by people with dementia appear to contribute to worse outcomes for people living with dementia. People with dementia have also been found to receive fewer palliative care treatments in hospital at the end of life than similar patients without dementia (Sampson et al., 2006).

With the negative effects of staying in hospital on people with dementia well-documented, many initiatives have focused on reducing the length of stay as much as possible. In the Netherlands, an initiative to reduce length of stay for people with dementia has adopted a “hospital at home” model for all but the most necessary treatment is in the process of being rolled out on a national level. Under this programme, people with dementia are admitted only for immediate treatment, such as surgery, with all possible follow-up conducted at home under the care of specialist nurses. Studies of hospital-at-home programmes have found that such programmes can lower the length of stay in hospital and reduce overall costs while delivering care of equal quality to acute hospital settings (Leff et al., 2005, Shepperd et al., 2009). For hospital at home to be effective, however, programmes may require intensive nursing and physician support, and strong communication and co-ordination links between community- and hospital-based care.

National audits of the hospital system can help to establish a baseline of care quality for dementia

Countries should consider undertaking national or regional reviews to help develop a baseline of the state of hospital care for people with dementia and identify key areas for improvement. In recent years, the United Kingdom, Ireland, and Switzerland have undertaken audits or investigations of hospital care quality for people with dementia, the results of which have fed into policy recommendations and changes. Results from the audits underscore how far hospitals still have to go in ensuring they deliver high-quality care for people with dementia. In Ireland, results of the 2013 hospital audit revealed that 94% of hospitals did not have a dementia care pathway in place, and that no hospitals required staff to undergo dementia awareness training (though such training was offered in about half of hospitals to doctors and nurses). 43% of patients had a mental status assessment noted in their health care records. In the United Kingdom, repeated audits of hospitals for dementia care quality have allowed them to track progress over time. They recently published the results of the third hospital dementia audit, which indicated that while progress has been made across a range of domains, further improvements can be made, including related to communication and information sharing, leadership, management of delirium, and person-centred care. National hospital audits have helped to inform policy development in both Ireland and the United Kingdom, and can serve as a model for other countries interested in improving hospital care for people with dementia.

4.2.2. Some countries are trying to improve the ability of hospital staff to effectively manage dementia

Staff training and key staff positions can help improve dementia care in hospitals

Studies of hospital staff indicate that the attitudes of health professionals, including nurses, significantly affect the care received by people living with dementia (Moyle et al., 2008; Dewing and Dijk, 2016). Many of the behavioural challenges of dementia, which can be exacerbated by the distress at being in hospital, can be difficult for hospital staff to manage effectively, and may contribute to higher levels of stress and burnout among staff members and poorer quality care, including unnecessary sedation, for people with dementia (Dewing and Dijk, 2016). Staff often report feeling under-trained in caring for dementia. This is particularly true among nurses and health care assistants, who spend more time with the patient than physicians, and may consequently impact the patient’s hospital experience more directly (Gandesha et al., 2012).

Given that hospital stays can be distressing for dementia patients, and can lead to lasting deterioration in their condition, hospitals in some OECD conditions have been looking for ways to improve the care experience. A number of countries have developed innovative new staff positions and teams working in hospitals to deal specifically with patients exhibiting symptoms of dementia. In Australia, the Dementia Behaviour Management Advisory Service is available in acute hospital settings to help deal with people with dementia exhibiting behavioural and psychiatric symptoms of dementia. In Ireland, a dementia nurse specialist role has been developed in acute hospital settings to serve as a link between hospital and community-based services. A similar role has been introduced in hospitals in Slovenia. Specialist mental health liaison teams in the United Kingdom serve a similar role, and help advise hospital staff on recognising and responding to delirium and dementia.

The emergence of new specialist roles to facilitate care for people with dementia may improve the quality of care received in hospital and help ease transitions between care settings. However, countries should be careful to introduce these new positions as an addition to, rather than replacement for, improving staff training in hospitals more broadly. These positions may often serve in a responsive crisis management role, rather than contributing to a broader shift in the approach to caring for people with dementia (Holmes et al., 2010; Dewing and Dijk, 2016).

Staff training initiatives for dementia available for hospital staff members are important in improving the hospital response to dementia. In Sweden, an internet-based training module targeted specifically at hospital staff was developed by the Swedish Dementia Centre, and has made the dementia certification programme available to hospitals in addition to other care units. In England, all hospital staff are required to undergo Tier 1 dementia awareness training according to the Dementia Core Skills Education and Training Framework, while staff members who more regularly interact with patients with dementia or direct services are required to have more advanced (Tier 2 and Tier 3) training on dementia. Very few countries reported a specific focus on improving the capacity of hospital staff to care for people with dementia. In most countries, initiatives have depended much more on the initiative of individual hospitals than country- or regional training programmes. Furthermore, even where training initiatives may be open to hospital staff, practical barriers, including cost, time away from work, and lack of professional incentives can hamper participation.

Delirium is a major driver of long hospital stays and poor outcomes and could be better managed

There is some evidence to suggest that interventions targeting delirium, frailty and other geriatric conditions that are not limited to people with dementia can have positive effects on the quality of care and outcomes for people with dementia in hospitals. Delirium is a major driver of longer lengths of stay in hospital, and many people with dementia will develop the condition upon entry into hospital (Ely et al., 2001; Fick et al., 2013). The presence of delirium is a strong predictor of worse outcomes among people with dementia (Morandi et al., 2014). Some estimates indicate that between 30-60% of elderly patients have both dementia and delirium while in hospital, meaning that interventions targeting delirium will impact a high proportion of people with dementia in hospital (Saravay et al., 2004). In some countries, including Switzerland, the United Kingdom, Australia, and New Zealand, national guidelines or in-hospital protocols for delirium mitigation have been developed. The presence of standardised protocols of care for people with dementia has been found to positively impact the quality of care received acute hospitals (Moyle et al., 2008).

Countries should encourage hospitals to develop training programmes and care pathways that help to minimise the prevalence and impact of delirium on patients. A number of models have been tested and validated in multiple health care systems, most notably the Hospital Elder Life Program intervention for preventing delirium and functional decline. HELP combines staff and volunteer training with a standardised intervention protocol and continual performance tracking mechanisms to ensure quality and compliance (Inouye et al., 2000; Rubin et al., 2011). Originally piloted in the United States, HELP has since been expanded to sites in eleven countries, including Australia, the Netherlands, Italy, and Japan (Hospital Elder Life Program, 2017).

Special care units can be helpful, but may not be able to treat all people with dementia needing hospital care

Special care units and geriatric hospital wards can offer older patients with dementia more highly specialised or intensive care than they might receive in a normal ward. Recognising the special care needs of older populations, many acute care hospitals have developed geriatric and sometimes special care units to care for elderly patients. Staff members working on geriatric wards are more highly trained in geriatric issues, including dementia and delirium. Special care units, which have been developed to care for people in hospital exhibiting behavioural symptoms of dementia or delirium, are staffed by personnel who have received additional training on responding to the behavioural symptoms of dementia and delirium. In England, most hospitals have elderly care wards that are designed to support people with dementia, while a number of hospitals have also developed dementia-specialist wards. Evaluations of special care units in Australia and Germany suggest that the care received can help to reduce negative outcomes of hospitalisation, including adverse events such as falls and reductions in capacity to perform activities of daily living (ADL), though they may not reduce the length of stay in hospital (Zieschang et al., 2010; Gonski and Moon, 2012). While geriatric and special care units may help to improve outcomes for those assigned to them, the sheer number of people admitted to hospital with dementia makes it unlikely that they can serve the entire population who could benefit. Countries must therefore continue to emphasise improvements in care quality across acute hospitals more generally, and not limit their efforts to only specialised care units.

4.2.3. Inconsistent coding practices are holding back quality of dementia care and the measurement of progress in hospitals

Even where effective care interventions are put in place, the impact on outcomes for people with dementia will be limited if dementia is not recognised upon admission. Recent work by the OECD on developing comparative dementia quality indicators suggests that under-coding of dementia remains a serious challenge in nearly all OECD countries. It is unlikely that countries can reliably identify all people with dementia in hospitals. Diagnoses of dementia made in other care settings are often not effectively transferred when the patient is admitted to hospital, leaving hospitals lacking a crucial piece of medical information that can inform their approach to care. Poor coding of dementia can also make it difficult to monitor and reduce preventable hospitalisations among people living with dementia, an objective of many country strategies to improve dementia care, such as Healthy People 2020 in the United States.

Furthermore, the low dementia diagnosis rate in the population overall suggests that many people visiting hospital may not be aware that they have dementia in the first place, or may not have received a formal diagnosis that would be recorded in their medical records. In these cases, a hospital visit offers an important opportunity for patients to receive an official diagnosis of dementia and be put on the pathway towards treatment and care. Physicians, nurses and other hospital staff should therefore be trained to recognise the signs of dementia, and care pathways in hospitals should be developed that can move patients quickly from detection to diagnosis.

There is a dramatic variation in the coding of dementia between countries which cannot be explained by differences in prevalence

The challenge of under-coding dementia is made clear by the gap in dementia diagnoses recorded in hospital admissions between the admissions-based and patient-based data reported as part of the 2017 data pilot undertaken by the OECD. Coding difficulties point to a significant challenge in using administrative datasets: when datasets are standalone and not linked to the patient across health care settings, administrative data is unlikely to offer a full portrait of dementia in countries. A patient presenting in the emergency room may have a dementia diagnosis recorded in their primary care or long-term care institution record, but if this information is not available to hospital staff upon evaluation and admission, they may not be recognised or recorded as having dementia. This can affect not only the accuracy of the administrative hospital data but also the quality of care the patient may receive. Regularly linking frequently used data sets can develop a more accurate picture of dementia. In Canada, for example, the Canadian Chronic Disease Surveillance System links hospital data, physician billing, drug data and health insurance registry, covering 97% of the population. It has tracked cases of dementia through the health system since 2002-2003.

It will also be challenging to use administrative data when overall diagnosis rates are low. For example, a country with low overall diagnosis rates may report lower-than-average admissions of people with dementia to hospital. Reducing hospital admissions among people with dementia is a worthy policy goal and can serve as an important indicator of quality of care. However, a low admissions rate in the context of low overall diagnosis rates may in fact indicate the opposite. Where overall diagnosis rates are low, the number of people recorded in hospital as having dementia will also be lower. Yet in these contexts, people with dementia likely receive less and possibly worse care than in health systems where concerted efforts to improve detection of dementia have led to more people receiving the care they need.

While admissions-based data identifies only people who had dementia recorded during a specific hospital visit, patient-linked data allows researchers to tie previous hospital visits to the same patient through the use of a unique patient identifier (UPI). This can help to identify more cases of patients who have received a diagnosis of dementia in the past, but who did not necessarily have dementia recorded on their latest hospital visit. Looking at hospital admissions which included a diagnosis of dementia as a proportion of all hospital admissions, up to a third of admissions included in the patient-based version of the indicator did not have a diagnosis of dementia recorded when they entered the hospital, even though they had previously been diagnosed. In France, for example, using patient-linked data resulted in the number of admissions with a diagnosis of dementia more than doubling. This indicates that there are a significant number of cases in which a diagnosis of dementia is not effectively communicated to the hospital, and is therefore not recorded in the hospital record.

Figure 4.6. Proportion of all-cause hospital admissions with dementia captured in the hospital record, where dementia was recorded in previous three years, 2015

Source: OECD Dementia Care Policy Interviews 2017.


Coding guidelines for dementia could help to improve coding rates and consistency in hospitals

Further efforts should be made to improve the consistency of dementia coding in acute care, both within countries (between hospitals and regions) and ultimately internationally. Without improved coding quality in hospitals, measuring and improving quality of care is nearly impossible, as differences can be driven as much by missed cases of dementia as by real differences in the quality of care. In many countries, coding guidelines have been developed that help coders to apply international disease codes, such as the ICD-9 and ICD-10 classification list. These guidelines can differ between countries and influence how dementia is recorded on patients’ medical records. In Australia, for example, dementia could previously only be coded (as either a principle or additional diagnosis) where dementia either caused the admission to hospital, or impacted care during the admission (Cummings et al., 2011). The Australian example also demonstrates how guidelines can impact dementia coding: the number of hospital admissions where dementia was recorded fell substantially following a change in coding standards in 2008. However, the decision of what constitutes an impact on care can be arbitrary, or influenced by what has previously been recorded in medical notes. More recently, a set of supplementary codes for chronic conditions, including dementia, were introduced in the 9th edition of the International Classification of Diseases, 10th revision, Australian Modification (ICD-10AM) and applies to all hospitalisation data reported since 1 July 2015. The Australian Coding Standard now states that dementia should be captured in the hospital record if present on admission, regardless of whether it was the reason for admission or required treatment. In the United Kingdom, coding guidelines were developed for general practitioners in London and the North of England after confusion over diagnostic codes was identified as a possible barrier to increasing the population in primary practice dementia registers (Russell et al., 2013). Countries should consider developing similar guidelines for applying dementia codes to hospital admissions. Doing so could help to increase the consistency in coding and ensure that possible cases of dementia do not go untreated.

Training hospital staff to recognise the symptoms of dementia may also help to improve detection and coding for dementia. It is possible that dementia is not only under-coded, but is also coded inconsistently, or only for a subset of people, such as those for whom dementia is the main reason for hospital care. Even where secondary diagnosis fields are available, physicians may not note a dementia diagnosis for an elderly patient unless accompanying symptoms, such as delirium, are present. More countries should consider adopting training interventions that improve the capacity of hospital staff working in the emergency department to recognise the symptoms of dementia, as is done in the Netherlands.

Countries are beginning to link datasets across settings, which could significantly improve the identification of dementia patients in hospitals

Of 22 OECD countries surveyed, slightly over half (13) report that they regularly link data from a minimum of four major national datasets (OECD, 2015b). However, important areas of the health system are largely excluded from data linking. Most importantly for dementia, very few countries link data systems from primary care settings (OECD, 2015b). Of 22 OECD countries, only two reported regularly linking primary care data to other pathways of health care. Including primary health data is particularly important for dementia patients given the central role primary care settings play in detecting, diagnosing, and caring for people with dementia. Where diagnosis is made by a primary care physician – as is the case for a majority of people with dementia in many countries – the exclusion of primary care from broader data linkages suggests that a patient’s diagnosis may not be regularly or accurately communicated when they go to access other parts of the health system.

Figure 4.7. Number of countries reporting national data used to conduct record-linkage projects on an occasional and on a regular basis

Source: OCED HCQI Questionnaire, Secondary Use of Health Data, 2011/12.


Countries must do more to improve hospital care for people with dementia

Many people with dementia will spend at least some time in an acute hospital, and it is important that hospitals are able to care for them in a way that minimizes the stress of a hospital setting. But in most countries, it is far from clear that hospitals can accurately identify all or even most people with dementia who seek care. The results of the pilot data analysis from the OECD demonstrates how in many countries, even people who had previously received a dementia diagnosis did not have this information recorded when they went to hospital. This can affect not only the accuracy of hospital data, but also the quality of care the patient receives. Coding standards that instruct health professionals on how to code diagnoses can impact how often diseases are recorded in hospital records. Because of the impact that dementia can have on even unrelated diseases, people who have been previously diagnosed with dementia should always have that information recorded in new care settings, and particularly when they visit acute hospitals. Countries should make sure that coding standards that do not impede recording dementia are in place.

4.3. Dying well: End of life and palliative care

Across OECD countries, people living with dementia continue to have insufficient access to palliative care at the advanced stages of the disease. In most countries, the palliative care model has been developed around care for cancer, a disease with a very different trajectory and end of life process than dementia. Because dementia is a terminal illness, much of the care process following a dementia diagnosis can, or should, adopt a palliative care approach, with an emphasis on person-centredness and delivering supportive rather than curative care. For the purposes of this section, however, the focus is more narrow, and examines how end-of-life palliative care is delivered to reduce pain and discomfort at the end-of-life stage of dementia (Box 4.4).

Box 4.4. What is palliative care?

According to the WHO, palliative care “is an approach that improves the quality of life of patients and their families facing the problem associated with life-threatening illness, through the prevention and relief of suffering by means of early identification and impeccable assessment and treatment of pain and other problems, physical, psychosocial and spiritual”(WHO, 2017).

Based on this and other definitions of palliative care, much of the care process for someone living with dementia can be considered to be providing palliative care, in the sense that care given to people with dementia is focused not on curing the disease, but in improving a person’s quality of life and comfort to the greatest extent possible for the rest of their life. For the purpose of this chapter, palliative care is used in the context of end of life palliative care to refer to the steps taken to maximise comfort and reduce pain at the advanced and end of life stages of dementia.

4.3.1. Advance care planning is important for people with dementia, but uptake remains low

Information about what advance care planning and palliative care options should be made available to people with dementia and their families and carers in a timely manner. In the majority of countries interviewed for this report, some form of advance care planning (ACP) is available to people living with dementia. Even where advance care planning services are technically available, however, uptake appears to be low. This continues to be the case even where palliative care is covered as part of the standard basket of health services, such as in Israel. Estimates of how often advance care planning is used are unavailable in most countries, though the first national prevalence study is currently underway in Australia, while in New Zealand, all persons who receive a home care package undergo an InterRAI assessment and have their advance care planning status recorded.

Many countries reported that while advance care planning can be introduced as part of post-diagnostic care, the frequency with which physicians and care workers raise the issue with people with dementia and their families is unclear. This is true even when advance care planning is part of clinical guidelines for dementia care, as is the case in Australia and the United Kingdom (England). Advanced directives for medical care have allowed people in the United Kingdom to refuse in advance specified medical treatments under the Mental Capacity Act (2005). However, the use of advanced directives is believed to be very low, and is estimated at about 4% in England, and 2% in Wales (YouGov, 2013).

Box 4.5. What is advance care planning?

Most broadly, advance care planning refers to discussion and planning of a person’s future living and care that takes place in anticipation of future physical or mental deterioration that would make communicating their wishes difficult. Some outcomes of advance care planning, including advance directives (AD) and power of attorney, are binding decisions that family members, caregivers, and health professionals are legally obligated to respect. Advance care planning may also entail non-legally binding decisions, including statements of wishes, or communication to family members and caregivers about what their future preferences would be. By helping to stimulate discussion and acceptance of the ultimate course of dementia, the advance care planning process is sometimes seen as more useful than any documents that may result (Sampson, 2010).

The decisions made during the advance care planning process are not necessarily legally binding. However, most OECD countries have developed legally-binding mechanisms that allow people with dementia to develop directives for care or appoint family members or others to serve as their financial or health care proxy (Table 4.1). In countries where legally binding advance directives are not available, physicians will have the final say on whether or not to respect the wishes conveyed by the person with dementia.

4.3.2. Palliative care for people with dementia poses specific challenges and countries should consider developing dementia-specific guidelines

Compared with cancer and other diseases, people with dementia have been found to receive notably less palliative care treatment at the end of life (Sampson et al., 2006). Patients with dementia receiving care in acute hospitals receive less palliative medication and fewer referrals to specialist palliative care than similar patients without a dementia diagnosis (Sampson et al., 2006). One study found that among patients who were admitted to hospital with a hip fracture, those with dementia received only one-third of the pain relief received by patients without dementia (Morrison and Siu, 2000).

Most notably, referrals to palliative care continue to be driven by cancer. In the United Kingdom (England, Wales and Northern Ireland), cancer caused 29% of deaths in 2012, but represented 88% of inpatient palliative care, and 75% of outpatient palliative care referrals (Dixon et al., 2015). Even as countries have begun making efforts to expand palliative care services to other conditions, cancer patients continue to be the primary beneficiaries of palliative care. Without reducing the availability of palliative care treatments for people with cancer, countries must make end-of-life palliative care more widely available to people living – and dying – with other disease.

While palliative care is available across the OECD, at least nine countries reported that it is unavailable or difficult to access for people with dementia. There is also a pressing need to make palliative care programmes more accommodating to the unique needs of people with dementia. In most countries, the development of palliative care systems has been based on the care trajectory of cancer. Providing palliative care to people with dementia can present different complications that should be taken into account. The inability of people with advanced dementia to communicate their level of pain can affect the pain relief they receive during their last months of life (Sampson, 2010). Difficulties in communicating their needs can drive people with dementia to exhibit behavioural challenges that can then result in further inappropriate care, such as the use of antipsychotics to control aggressive behaviour (OECD, 2015a). Even where specialists have been trained in palliative care, they may be less familiar with the complexities of the end of life in people with dementia.

Table 4.1. Availability of legally-binding advance care planning mechanisms


Advance directive

Power of attorney (legal proxy)










Czech Republic








Yes (financial proxy only)












Yes (financial proxy only)












Yes (financial proxy only)








Yes (12 states + Mexico City)





New Zealand





Yes (financial proxy only)







Slovak Republic








Yes (financial proxy only)



Yes (financial proxy only)






Yes (health care proxy only)

United Kingdom



United States



Source: OECD Dementia Care Policy Interviews 2017; Alzheimer Europe 2017.

The development of guidelines outlining the use of palliative care among people with dementia has been demonstrated to increase the use of palliative care. The implementation of multidisciplinary guidelines on palliative care for people with dementia in the United Kingdom significantly reduced the use of antibiotics and increased the administration of pain relief during the last two weeks of life in a psychiatric hospital (Lloyd-Williams and Payne, 2002). Care staff and health workers may be unclear about the symptoms of dying in people with dementia, or uncertain about how palliative care should be applied when people are unable to communicate or clearly demonstrate they are in pain. The development of guidelines and frameworks for administering palliative care to people with dementia could help to streamline and institutionalise the use of palliative care at the end of life. No countries have developed national frameworks for administering palliative care to people with dementia, though a limited number have identified people with dementia as a target population for increasing access to palliative care.

The availability of palliative care services varies widely between countries, though in nearly all cases, the majority of palliative care places are located in tertiary hospitals. In some countries, including Denmark, Mexico and Slovenia, palliative care is not available to people with dementia who live at home. Providing palliative care at home or in a familiar environment may cause less distress for dementia sufferers and their carers and families.

In the United Kingdom, the government has committed to promoting high quality and personalised end of life care, including end of life care delivered at home and in care facilities. To improve end of life care for residents of care facilities, best practice and key lessons have been disseminated through an Enhanced Health in Care Homes framework, as well as a series of best practices published by Public Health England in 2017. Guidance published in May 2017 focuses specifically on how care staff should approach care planning for people with dementia, including examples of best practice.

In some cases, payment systems have created unnecessary barriers to accessing palliative care for people with dementia. Reimbursement practices in health systems should be reviewed to ensure that people with dementia can have end-of-life services reimbursed in a straightforward manner. In the United States, hospice care is only accessible upon receiving a prognosis of dying within six months, with only 30 days of hospice reimbursed by Medicare. People with dementia may cycle multiple times through hospice care and nursing homes as they live beyond their initial prognosis and overstay the hospice period funded by Medicare. In other countries, including Germany and the Netherlands, dementia may not always be classified as a terminal disease, creating payment challenges at the end of life when services designated for terminal patients – including palliative care and hospice – are needed (Davies et al., 2014).

4.3.3. Helping people with dementia at the end of life

The difficulties in understanding and communicating that accompany advanced stages of dementia makes meeting the needs of people with dementia at the end of life particularly difficult. Promoting advance care planning options that encourage difficult end-of-life discussions to occur before people with dementia have lost their cognitive function and ability to communicate can help them to convey their preferences before it is too late. Expanding palliative care programmes is also important to ensuring people with dementia do not experience unnecessary pain or discomfort at the end of life. Countries should ensure that existing palliative care systems adequately cover people with dementia. Where this is not the case, palliative care policies, including reimbursement policies and the location of palliative care services, should be reconsidered to ensure people with dementia at the end of life can access the services they need.


Auer, S. et al. (2017), “Wohngemeinschaft versus traditionelles Pflegeheim für Personen mit Demenz”, pp. 1–10.

Ayalon, L. et al. (2006), “Effectiveness of Nonpharmacological Interventions for the Management of Neuropsychiatric Symptoms in Patients With Dementia”, Archives of Internal Medicine, Vol. 166, No. 20, pp. 2182–2187.

Björk, S. et al. (2016), “Exploring the prevalence and variance of cognitive impairment, pain, neuropsychiatric symptoms and ADL dependency among persons living in nursing homes; a cross-sectional study”, BMC Geriatrics, Vol. 16, No. 1.

Boustani, M. et al. (2005), “Characteristics Associated With Behavioral Symptoms Related to Dementia in Long-Term Care Residents”, The Gerontologist, Vol. 45, No. Special Issue 1, pp. 56–61,

Chenoweth, L. et al. (2009), “Articles Caring for Aged Dementia Care Resident Study (CADRES) of person-centred care, dementia-care mapping, and usual care in dementia: a cluster-randomised trial”, The Lancet Neurology, Vol. 8, No. 4, pp. 317–325.

Commings, E. et al. (2011), “Hospital coding of dementia: is it accurate?”, Health Information Management Journal, Vol. 40, No. 3, pp. 5-11.

Davies, N. et al. (2014), “Quality palliative care for cancer and dementia in five European countries: some common challenges”, Aging & Mental Health, Vol. 18, No. 4, pp. 400-410,

Dewing, J. and S. Dijk (2016), “What is the current state of care for older people with dementia in general hospitals? A literature review”, Dementia, Vol. 15, No. 1, pp. 106–124.

Dixon, J. et al. (2015), “Equity in the Provision of Palliative Care in the UK: Review of Evidence” (pp. 1–148). London: Personal Social Services Research Unit, London School of Economics and Political Science.

Ely, E. et al. (2001), “The impact of delirium in the intensive care unit on hospital length of stay”, Intensive Care Medicine, Vol. 27, No. 12, pp. 1892–1900.

Emilsson, L. et al. (2015), “Review of 103 Swedish Healthcare Quality Registries”, Journal of Internal Medicine, Vol. 277, No. 1, pp. 94–136.

Fick, D. M. et al. (2013), “Delirium superimposed on dementia is associated with prolonged length of stay and poor outcomes in hospitalized older adults”, Journal of Hospital Medicine, Vol. 8, No. 9, pp. 500–505.

Gandesha, A. et al. (2012), “Adequacy of training in dementia care for acute hospital staff”, Nursing Older People, Vol. 24, No. 4, pp. 26-31.

Gonski, P. N. and I. Moon (2012), “Outcomes of a behavioral unit in an acute aged care service”, Archives of Gerontology and Geriatrics, Vol. 55, No. 1, pp. 60–65.

Gulpers, M. J. M. et al. (2011), “Belt Restraint Reduction in Nursing Homes: Effects of a Multicomponent Intervention Program”, Journal of the American Geriatrics Society, Vol. 59, No. 11, pp. 2029–2036.

Halfens, R.J.G. et al. (2013), “International prevalence measurement of care problems: results”, Journal of Advanced Nursing, Vol. 69, No. 9,

Hamers, J. P. H. et al. (2009), “Attitudes of Dutch, German and Swiss nursing staff towards physical restraint use in nursing home residents, a cross-sectional study”, International Journal of Nursing Studies, Vol. 46, No. 2, pp. 248–255.

Holmes, J. et al. (2010), Liaison Mental Health Services for Older People: A Literature review, service mapping and in-depth evaluation of service models, National Institute for Health Research Service Delivery and Organisation programme.

Hospital Elder Life Program (2018), “Hospital elder life program (HELP) for prevention of delirium”, [website],

Hughes, J. et al. (2008), “Care staff working with people with dementia: Training, knowledge and confidence”, Dementia, Vol. 7, No. 2, pp. 227–238.

Inouye, S. K. et al. (2000), “The Hospital Elder Life Program: A Model of Care to Prevent Cognitive and Functional Decline in Older Hospitalized Patients”, Journal of the American Geriatrics Society, Vol. 48,, pp. 1697–1706.

Kane, R. A. et al. (2007), “Resident Outcomes in Small-House Nursing Homes: A Longitudinal Evaluation of the Initial Green House Program”, Journal of the American Geriatrics Society, Vol. 55, No. 6, pp. 832–839.

Koczy, P. et al. (2011), “Effectiveness of a Multifactorial Intervention to Reduce Physical Restraints in Nursing Home Residents”, Journal of the American Geriatrics Society, Vol. 59, No. 2, pp. 333–339.

Köpke, S. et al. (2012), “Effect of a Guideline-Based Multicomponent Intervention on Use of Physical Restraints in Nursing Homes”, The Journal of the American Medical Association, Vol. 307, No. 20, pp. 2177–2184.

Leff, B. et al. (2005), “Hospital at Home: Feasibility and Outcomes of a Program To Provide Hospital-Level Care at Home for Acutely Ill Older Patients”, Annals of Internal Medicine, Vol. 143, No. 11, pp. 798–808.

Livingston, G. et al. (2005), “Systematic Review of Psychological Approaches to the Management of Neuropsychiatric Symptoms of Dementia”, American Journal of Psychiatry, Vol. 162, No. 11, pp. 1996–2021.

Lloyd-Williams, M. and S. Payne (2002), “Can multidisciplinary guidelines improve the palliation of symptoms in the terminal phase of dementia?”, International Journal of Palliative Nursing, Vol. 8, No. 8, pp. 371–375.

Lum, T. Y. et al. (2009), “Effects of Green House Nursing Homes on Residents' Families”, Health Care Financing Review, Vol. 30, No. 2, pp. 35–51.

Morandi, A. et al. (2014), “Delirium Superimposed on Dementia Strongly Predicts Worse Outcomes in Older Rehabilitation Inpatients”, Journal of the American Medical Directors Association, Vol. 15, No. 5, pp. 349–354.

Morrison, R.S. and A.L. Siu (2000), “A Comparison of Pain and Its Treatment in Advanced Dementia and Cognitively Intact Patients with Hip Fracture”, Journal of Pain and Symptom Management, Vol. 19, No. 4, pp. 240–248.

Morrison, R.S. and A.L. Siu (2000), “Survival in end-stage dementia following acute illness”, Journal of the American Medical Association, Vol. 284, No. 1, pp. 47-52.

Moyle, W. et al. (2008), “Best practice for the management of older people with dementia in the acute care setting: a review of the literature”, International Journal of Older People Nursing, Vol. 3, No. 2, pp. 121–130.

Nakahira, M. et al. (2009), “Attitudes toward dementia-related aggression among staff in Japanese aged care settings”, Journal of Clinical Nursing, Vol. 18, No. 6, pp. 807–816.

Nightingale, S. et al. (2001), “Psychiatric illness and mortality after hip fracture”, The Lancet, Vol. 357, No. 9264, pp. 1264–1265.

OECD (2015a), Addressing Dementia: The OECD Response, OECD Health Policy Studies, OECD Publishing, Paris,

OECD (2015b), Health Data Governance: Privacy, Monitoring and Research, OECD Health Policy Studies, OECD Publishing, Paris,

Onder, G. et al. (2012), “ Assessment of nursing home residents in Europe: the Services and Health for Elderly in Long TERm care (SHELTER) study”, BMC Health Services Research, Vol. 12, No. 5.

PHI. (2017). “Workforce Data Centre”,

Rubin, F. H. et al. (2011), “Sustainability and Scalability of the Hospital Elder Life Program at a Community Hospital”, Journal of the American Geriatrics Society, Vol. 59, No. 2, pp. 359–365.

Russell, P. et al. (2013), “Improving the identification of people with dementia in primary care: evaluation of the impact of primary care dementia coding guidance on identified prevalence”, BMJ Open, Vofl. 3, No. 12.

Sampson, E. L. et al. (2006), “Differences in care received by patients with and without dementia who died during acute hospital admission: a retrospective case note study”, Age and Ageing, Vol. 35, No. 2, pp. 187–189.

Sampson, E. L. et al. (2009), “Dementia in the acute hospital: prospective cohort study of prevalence and mortality”, The British Journal of Psychiatry, Vol. 195, No. 1, pp. 61–66.

Sampson, E. L. (2010), “Palliative care for people with dementia”, British Medical Bulletin, Vol. 96, No. 1, pp. 159–174.

Saravay, S. M. et al. (2004), “How Do Delirium and Dementia Increase Length of Stay of Elderly General Medical Inpatients?”, Psychosomatics, Vol. 45, No. 3, pp. 235–242.

Shepperd, S. et al. (2009), “Avoiding hospital admission through provision of hospital care at home: a systematic review and meta-analysis of individual patient data”, Canadian Medical Association Journal, Vol. 180, No. 2, pp. 175–182.

Smyth, W. et al. (2013), “A survey-based study of knowledge of Alzheimer’s disease among health care staff”, BMC Geriatrics, Vol. 13, No. 2.

Spector, A., Orrell, M. and J. Goyder (2013), “A systematic review of staff training interventions to reduce the behavioural and psychological symptoms of dementia”, Ageing Research Reviews, Vol. 12, No. 1, pp. 354–364.

Sweden National Quality Registers, (2016), National Quality Registry for Behavioural and Psychological Symptoms of Dementia (BPSD), Nationella Kvalitetsregister.

Teri, L. et al. (2005), “STAR: A Dementia-Specific Training Program for Staff in Assisted Living Residences”, The Gerontologist, Vol. 45, No. 5, pp. 686–693.

Testad, I., Aasland, A. M. and D. Aarsland (2005), “The effect of staff training on the use of restraint in dementia: a single-blind randomised controlled trial”, International Journal of Geriatric Psychiatry, Vol. 20, No. 6, pp. 587–590.

Verbeek, H. et al. (2010), “Dementia Care Redesigned: Effects of Small-Scale Living Facilities on Residents, Their Family Caregivers, and Staff”, Journal of the American Medical Directors Association, Vol. 11, No. 9, pp. 662–670.

Verbeek, H. et al. (2014), “Effects of small-scale, home-like facilities in dementia care on residents’ behavior, and use of physical restraints and psychotropic drugs: a quasi-experimental study”, International Psychogeriatrics, Vol. 26, No. 04, pp. 657–668.

World Health Organization. (2018), WHO Definition of Palliative Care [website], retrieved from

Zieschang, T. et al. (2010), “Improving care for patients with dementia hospitalized for acute somatic illness in a specialized care unit: a feasibility study”, International Psychogeriatrics, Vol. 22, No. 1, pp. 139–146.