Shorter Oxford Textbook of Psychiatry, 6th Ed.

CHAPTER 18. Psychiatry of the elderly


Normal ageing

Epidemiology of psychiatric disorders in the elderly

Principles and practice of old age psychiatry

Treatment of dementia

Clinical features and treatment of psychiatric disorders in the elderly


Older people with mental health problems present particular challenges to the practice of old age psychiatry and the organization of its services. They are often physically as well as mentally frail, and this affects presentation and course. On the other hand, they have the advantage of a rich history to tell and a lifetime’s experience of responding to fortune and adversity. Dementia comprises a substantial part of the clinical practice.

When considering psychiatric disorder in the elderly, the clinician must be able to collect and integrate information from a variety of sources, and produce a management plan which takes account of physical and social needs, as well as psychological ones. This plan is likely to involve the cooperation of several professionals. It is in this clinical complexity that much of the challenge and fascination of old age psychiatry lies.

This chapter deals with the psychiatry of old age, with two important exceptions, both of which were covered in Chapter 13:

• delirium (see pp. 317–19)

• the clinical features, aetiology, and investigation of dementia (see pp. 321–35).

Normal ageing


In 1993, 6% of the world’s population was over 65 years of age. However, in the more developed countries the proportion was about 14%, while in less developed countries it was 4%. Less developed countries have higher birth rates, but the life expectancy at birth is substantially lower than in developed countries—60 years, compared with 73 years. In the UK, life expectancy at birth has increased from 41 years in 1840 to 46 years in 1900, 69 years in 1950, and 80 years in 2011.

Table 18.1 shows that the difference in age structure of the population is changing, and Table 18.2 shows that the proportion of older people in less developed countries is increasing much faster than it is in developed countries. Although some psychiatric disorders become less common (see Figure 18.1), the prevalence of dementia increases rapidly with age (see below), and all countries face the increasing problem of managing large numbers of cognitively impaired older people.

Table 18.1 Percentage change in the world population during the period 1975–2000


Physical changes in the brain

The weight of the human brain decreases by approximately 5% between the ages of 30 and 70 years, by a further 5% by the age of 80 years, and by another 10–20% by the age of 90 years. As well as these changes, the ventricles enlarge and the meninges thicken. MRI studies show a complex temporal and spatial profile of changes affecting both grey and white matter, with volume reductions prominent in the hippocampus, frontal cortex, and cerebellum (Caserta et al., 2009).


Figure 18.1 Prevalence of mental disorders across age groups: data are 1-month prevalence rates from the Epidemiologic Catchment Area Study using DSM-III criteria. From Jorm (2000).

Table 18.2 Percentage of the population aged 60 years or over, 65 years or over, and 80 years or over by gender


There is some loss of neurons, although this is regionally selective and much less marked than was formerly believed, and reductions in synapses and dendrites appear to be more important. The cytoplasm of some neurons contains a pigment, lipofuscin. The ageing brain also tends to accumulate senile plaques and neurofibrillary tangles, but with a more restricted distribution and smaller numbers than in Alzheimer’s disease (see Chapter 13). Neurofibrillary tangles are usually limited to neurons in the hippocampus and entorhinal cortex, while senile plaques can also occur in the neocortex and amygdala. Similarly, a small proportion of brains from healthy old people contain Lewy bodies. For reviews of the neuropathology of normal ageing, see Hof and Morrison (2004) and Yankner et al. (2008).

Ageing itself is thought to reflect genomic changes (e.g. acquired damage to DNA or its epigenetic regulation), mitochondrial damage (due in part to disorders of calcium regulation and to free radicals), and alterations in some growth factor and signalling pathways, together with the accumulation of multiple random (stochastic) changes. For a review of the biology of ageing, see Bittles (2009).

The neuropsychology of ageing

Assessment of cognitive function in the elderly is complicated by the frequent presence of physical ill health, notably sensory deficits, and by the need to carefully distinguish normality from the earliest phase of dementia.

Longitudinal studies suggest that intellectual function, as measured by standard intelligence tests, shows a significant decline only in later old age. A characteristic pattern of change occurs, with psychomotor slowing and impairment in the manipulation of new information. By contrast, tests of well-rehearsed skills such as verbal comprehension show little or no age-related decline.

Short-term memory, as measured by the digit span test, for example, does not change in the normal elderly. Tests of working memory show a gradual decrease in capacity, so that the elderly perform significantly less well than the young if attention has to be divided between two tasks or if the material has to be processed additionally in some way. The elderly can usually recall remote events of personal significance with great clarity. Despite this, their long-term memory for other remote events shows a decline. Overall, there appears to be a balance between losses in flexible problem solving, and the benefits of accumulated wisdom derived from experience. For a review, see Anderson (2008).

As well as these cognitive and motor changes, there are alterations in personality and attitudes, such as increasing cautiousness and rigidity.

Physical health

In addition to a general decline in functional capacity and adaptability with ageing, chronic degenerative conditions are common. As a result, the elderly consult their family doctors frequently and occupy around 50% of all hospital beds. These demands are particularly large in those aged over 75 years. Medical management is made more difficult by the presence of more than one disorder, the consequent increased risk of side-effects of treatment, and by psychiatric and social problems.

Sensory and motor disabilities are frequent among elderly people. In a study of people over 80 years of age in Jerusalem, 50–60% reported problems with vision, hearing, or walking, and 22% had difficulties in talking (Davies and Fleishman, 1981). These figures are representative of those reported in other countries.

Social circumstances

For most people, ageing brings with it profound changes in social circumstances. Retirement affects not only income but also social status, time available for leisure, and social contacts. Loss of income is a serious problem facing many elderly people, and financial problems were the commonest worry reported in a large European survey of people over 65 years of age.

Social isolation is a fact of life for many older people, especially in developed countries. In the UK and the USA, about one-third of those over 65 years of age were living alone in 1993, compared with 7% in Chile and 3% in China. The figure in the UK increased markedly in the second half of the last century (by contrast, in 1851 it was 7%, and in 1921 it was 11%). However, for many older people, living alone is not seen as a problem and is aided by slowly increasing pensioner incomes and greater availability of suitable housing. Many older people see family, friends, and neighbours regularly, and provide as much support as they receive. In a European Commission survey in 1993, 44% of older Europeans saw a relative every day. This figure was highest in southern Europe and lowest in northern Europe, but, interestingly, reported loneliness showed the same pattern, being highest in countries with the highest levels of family contact.

Table 18.3 Percentages of people living alone, or in residential care, in England and Wales in 2001


Table 18.3 summarizes data from the 2001 Great Britain Census ( on the proportion of elderly people who live at home alone, or who live in a ‘communal establishment’ (i.e. residential care or long-stay hospital bed). In both sets of statistics, women predominate, reflecting the greater number of widows than widowers (and thus women who lack a partner to care for them), and the higher level of disability reported by women than men at any given age. Note also that the lower row indicates that 93% of elderly people, including 67% of those over 85 years of age, live in their own home rather than in residential care, contrary to popular belief.

Epidemiology of psychiatric disorders in the elderly

Kay et al. (1964) conducted the first systematic prevalence study of psychiatric disorder among elderly people in the general population, including those living at home as well as those living in institutions, in an area of Newcastle upon Tyne, in northern England. The findings (see Table 18.4) have been broadly replicated in subsequent surveys, taking into account changes in diagnostic criteria (e.g. inclusion of depression in their category of neurosis), and the problems of case finding. For example, a large French population study of people aged 65 years or over found a point prevalence of 14% for anxiety disorder, 11% for phobia, 3% for major depression, and 1.7% for psychosis, using an interview to make DSM-IV diagnoses (Ritchie et al., 2004). These figures mask the significant changes in the prevalence and proportions of disorders at different stages of old age, as shown by a large Swedish study (Skoog, 2004), summarized in Table 18.5.

Other surveys have shown a high prevalence of psychiatric disorder among elderly people in sheltered accommodation and in hospital. One-third of the residents in old people’s homes have significant cognitive impairment. In general hospital wards, between one-third and half of the patients aged 65 years or over have some form of psychiatric illness.

Table 18.4 Prevalence of psychiatric disorder in people over 65 years of age


Table 18.5 Changing prevalence of psychiatric disorders with age among the elderly


It has frequently been reported that general practitioners are unaware of many of the psychiatric problems among elderly people living in the community. Moreover, the presentation of such disorders to general practitioners and psychiatrists is determined as much by social factors as by a change in the patient’s mental state. For example, there may be a sudden alteration in the patient’s environment, such as illness of a relative, or a bereavement. Sometimes an increasingly exhausted or frustrated family decides that they can no longer continue to care for an old person.

As in younger people, there are gender differences in the prevalence, presentation, and course of some psychiatric disorders, and in treatment needs (Lehmann, 2003).

For a review of the epidemiology of psychiatric disorders in the elderly, see Henderson and Fratiglioni (2009).


Although there are references in classical literature, dementia in the elderly has been recognized by modern medicine since the French psychiatrist Jean-Étienne Esquirol described démence sénile in 1838. Esquirol’s description was in general terms, but it can be recognized as similar to the present-day concept (Alexander, 1972). Emil Kraepelin distinguished dementia from psychoses due to other organic causes such as neurosyphilis, and he divided it into presenile, senile, and arteriosclerotic forms. In an important study, Roth (1955) showed that dementia in the elderly differed from affective disorders and paranoid disorders in its poorer prognosis.

Following Roth (1955), there has been extensive research on the prevalence of dementia, and several meta-analyses (see Figure 18.2 and Table 18.6). For a review, see Ferri et al. (2005) and Reitz et al. (2011). Prevalence rises steeply with age, up to the age of 90 years, and thereafter the increase slows, and about 25% of centenarians remain cognitively intact. As discussed in Chapter 13, the commonest causes are Alzheimer’s disease, vascular dementia, and dementia with Lewy bodies, with frequent coexistence of the features of more than one disorder.


Delirium and its many causes were discussed in Chapter 13. Age is also a major risk factor. For example, in the community, rates of delirium rise from 1–2% to at least 14% in those over 85 years of age (Rahkonen et al., 2001). Elderly patients in hospital are at particular risk, with about one-third experiencing an episode of delirium (see Chapter 13).

Mood disorder

Despite the focus on dementia in the elderly, depressive disorders are considerably more common. A systematic review of community-based studies found an average prevalence of clinically relevant depression of 13.5% for those aged 55 years or over, of which 9.8% was classed as minor and 1.8% as major. Prevalence was higher in women and among older people living in adverse circumstances (Beekman et al., 1999). Consistent with these data, a large European collaborative study found rates of 8.5–14% for clinically significant depression in the elderly and 1–4% for major depression (Copeland et al., 1999). The figures show that a high proportion of depressed elderly people do not meet the criteria for major depression, but are instead variously labelled as having minor depression, dysthymia, or subthreshold depression. However, this does not diminish its clinical significance, as these cases have a similar morbidity and course to those of patients who meet the criteria for major depression (Beekman et al., 2002).


Figure 18.2 Prevalence rates for dementia across age groups: data from three meta-analyses. Reproduced from A. R. Lingford-Hughes, S. Welch, D. J. Nutt, Journal of Psychopharmacology, 18 (3), Evidence-Based Guidelines for the Pharmacological Management of Substance Misuse, Addiction and Comorbidity: Recommendations from the British Association for Psychopharmacology, pp. 293–335, copyright © 2004 by (Sage Publications). Reprinted by Permission of SAGE.

Table 18.6 Prevalence and incidence of dementia in different populations


Rates of depression depend on the setting. Koenig and Blazer (1992) reported rates of 0.4–1.4% in the community, 5–10% among medical outpatients, 10–15% among medical inpatients, and 15–20% among nursing home patients. These variations probably reflect the frequent comorbidity of depression in the elderly with other psychiatric (Devanand, 2002) and physical (Krishnan et al., 2002) disorders. For example, rates of depression are higher in those with cardiac problems and neurodegenerative disorders, and one-third of patients have an alcohol misuse disorder. These comorbidities may also contribute to the increased disabilities and mortality rate, especially from cardiac events (Penninx et al., 1999), found in elderly people with depression. Elderly prisoners have particularly high rates of depression (Fazel et al., 2001).

The high prevalence of depressive disorder masks the fact that first episodes become less common after the age of 60 years, and rare after the age of 80 years. A new onset of mania or bipolar disorder in the elderly is very rare, and an organic cause (e.g. secondary to steroids) should always be suspected (Depp and Jeste, 2004).

The incidence of suicide, especially in men, increases steadily with age (see Chapter 16), and suicide in the elderly is usually associated with depressive disorder (Waern et al., 2002).

Anxiety disorders

In general practice, Shepherd et al. (1966) found that, after the age of 55 years, the incidence of new cases of neurosis (i.e. anxiety disorder) declined. However, the frequency of consultations with the general practitioner for neurosis did not fall—presumably as a result of chronic or recurrent cases. Subsequent larger population surveys have largely confirmed this view, although prevalence and incidence figures differ markedly between studies depending on the case definitions employed. For example, in a study from Liverpool, 1-month prevalence rates for ‘caseness’ for neurotic disorders were less than 1% in men and less than 2% in women over 65 years of age, whereas ‘subcase’ rates were about 20% in both sexes. In a 3-year follow-up, this corresponded to an incidence rate of 4.4 per 1000 per year. For a review, see Lindesay (2008).

Schizophrenia-like disorders

Schizophrenia-like and paranoid disorders in the elderly have been a long-standing source of debate and terminological confusion, as discussed in Chapter 12 (see Box 12.2), with the terms paraphrenia and late paraphrenia still commonly, if loosely, used. In future versions of ICD and DSM, the cumbersome term ‘very-late-onset schizophrenia-like psychosis’ will probably cover psychoses of this kind that arise after 60 years of age. The delineation of this group, and its separation from ‘late-onset schizophrenia’, which will apply to those with onset between the ages of 40 and 59 years, is based on differences in its symptom profile and associated risk factors (Howard et al., 2000).

Patients with ‘very-late-onset schizophrenia-like psychosis’ account for approximately 10% of admissions to psychiatric wards for the elderly. There are no reliable data on the prevalence in the community, as it is difficult to identify all cases of a condition in which many sufferers keep their experience to themselves and are unlikely to cooperate with ‘doorstep’ interviews. Also, in this setting, it may be difficult to distinguish cases from those with psychotic symptoms occurring in dementia or in severe depression.

Principles and practice of old age psychiatry

Having surveyed the epidemiological landscape, in this section we outline the principles and practice of old age psychiatry, in terms of service organization, assessment, treatment, and legal issues. The remainder of the chapter describes the specific features and treatments of the individual psychiatric disorders of old age.

Organization of services

An international consensus statement defined the essential elements of a mental health service for elderly people as follows (Wertheimer, 1997):

• primary healthcare team

• specialist old age psychiatry team

• inpatient unit

• rehabilitation

• day care

• availability of respite care

• range of residential care facilities

• family and social supports

• liaison with geriatric medicine

• education of healthcare providers about the needs of elderly people with psychiatric problems

• research, especially into epidemiological issues, and evaluation of services.

The ‘team’ concept is highlighted as being the key to effective provision of care (Banerjee and Chan, 2008), with multidisciplinary involvement being the norm. However, beyond these broad points of consensus, national policies have differed considerably in terms of the development of services and how they are implemented (Johnston and Reifler, 2000). In the USA, emphasis has been placed on care in hospitals and nursing homes. In Europe, Canada, and Australasia, there has been varying emphasis on social policies to provide sheltered accommodation and care in the community. In this section, services in the UK will be described as an example. Health, social, and voluntary services will be described separately, although good care depends on close collaboration at all levels, from strategic planning to the coordinated provision of care for each individual patient. Reflecting this need, health and social services, and their budgets, are now usually integrated in the UK. A National Dementia Strategy, ‘Living well with dementia’ (Department of Health, 2009), has been launched for England and Wales. It proposes three key steps:

1. Ensure better knowledge about dementia and remove stigma.

2. Ensure early diagnosis, support, and treatment for people with dementia and their families and carers.

3. Develop services to better meet changing needs.

Health services

Primary care

The general practitioner has a central role in assessment and management of the problems of mentally ill older people. In the UK, all patients over the age of 75 years are offered an annual health check, providing an opportunity for regular screening for psychiatric disorders. In many instances, the general practitioner, together with other health professionals in the primary care team and community health services, assesses and manages patients without referring them to a specialist. Even in the case of disorders for which referral will usually follow, such as dementia, the general practitioner is still responsible for the initial assessment and diagnosis (Turner et al., 2004). However, as already mentioned, general practitioners do not detect all of the psychiatric problems of the elderly at an early stage, nor do they always provide all of the necessary long-term medical supervision. These problems are due partly to lack of awareness of the significance of psychiatric illness among the elderly, and partly to a traditional service model in which doctors respond to requests from patients rather than seeking out their problems. Some old age psychiatry services will only accept referrals from a general practitioner, but increasingly referrals may be accepted directly from any source, although liaison with the general practitioner will always be important. In some areas, community psychiatric nurses (CPNs) work in primary care, although it is more usual for them to be part of old age psychiatry teams in secondary care.

Old age psychiatry services

The organization of psychiatric services varies in different localities, as it reflects the local styles of service providers, local needs, and the extent of provision for this age group by general psychiatric services, as well as national policies. Nevertheless, there are some general principles of planning. The aims should be to maintain the elderly person at home for as long as possible, to respond quickly to medical and social problems as they arise, to ensure coordination of the work of those providing continuing care, and to support relatives and others who care for the elderly person at home. There should be close liaison with primary care, other hospital specialists who may be involved, social services, and voluntary agencies. A multidisciplinary approach should be adopted with a clinical team that may include psychiatrists, psychologists, community psychiatric nurses, occupational therapists, and social workers. Some members of the team should expect to spend more of their working day in patients’ homes and in general practices than in the hospital. It is important that the care provided by these different agencies is coordinated, and in the UK the Care Programme Approach is intended to improve coordination by making a single person, the key worker or ‘care-coordinator’, responsible for ensuring that appropriate assessment, care planning, and review take place (see p. 607).

The contributions of the various parts of an old age psychiatry service will now be considered. In England and Wales, current severe financial constraints and the reorganization of the NHS to move control of budgets to general practitioners mean that the description of services for older people given below will inevitably change over the lifetime of this edition of this book. Although the basic principles and components of care may not change radically, the relative availability of these components may do so, and the reader will need to familiarize him- or herself with local conditions.

Domiciliary psychiatric care. Increasingly, assessment and treatment take place in the patient’s own home, which is both more convenient for the patient, and offers a more relevant and realistic assessment of the difficulties facing the patient and their carers. Community psychiatric nurses act as a bridge between primary care and specialist services. The nurses may assess referrals from general practitioners, monitor treatment in collaboration with general practitioners and the psychiatric services, and take part in the organization of home support for the demented elderly.

Outpatient clinics. These have a smaller part to play in providing care for the elderly than in providing it for younger patients, because assessment at home is particularly important for old people. However, such clinics are convenient for the assessment and follow-up of mobile patients. There are advantages when these clinics are staffed jointly by medical geriatricians and old age psychiatrists. In recent years, memory clinics have been developed in most areas for the specialist assessment of patients with early memory problems (Kelly, 2008). In the UK, these clinics also serve as the focus for prescription of drugs used to treat Alzheimer’s disease, since their availability is limited (see p. 500).

Day hospitals and day centres. Although some treatments can be provided at home, others may require the patient to attend a day hospital, where a high level of stimulation and social interaction can also be provided. In the 1950s, day care began in geriatric hospitals. A few years later the first psychiatric day hospitals for the elderly were opened. Psychiatric day hospitals should provide a full range of diagnostic services and offer both short-term and continuing care for patients with functional or organic disorders, together with support for relatives. Currently, day care is more often provided by day centres rather than in day hospitals. These centres combine health and social care elements, very often with the financial and practical involvement of voluntary organizations, such as the Alzheimer’s Society. Day centres provide an important and cost-effective component of the services available for the elderly with psychiatric problems. Day care has become a mainstay of care for those with mild to moderate dementia. All day-care-based arrangements depend crucially on adequate transport facilities.

Inpatient units. Inpatient teams should be able to provide multidisciplinary assessment and treatment of patients with severe disorders. In addition to psychiatrists and psychiatric nurses, these teams may include occupational therapists, psychologists, speech and language therapists, physiotherapists, social workers, healthcare assistants, and others. There is substantial variation in different areas both in the composition of the team and in whether patients with functional illnesses are cared for separate from or together with those with organic disorders.

Geriatric medicine. There is inevitably some overlap in the characteristics of patients treated by units for geriatric medicine and those treated in psychiatry units; both are likely to treat patients with dementia. However, old age psychiatry units are more likely to treat patients with functional psychiatric disorders. Earlier concern that many patients were ‘misplaced’ and therefore received poor treatment, stayed too long in hospital, and had an unsatisfactory outcome has not been confirmed by research. Copeland et al. (1975) found that although 64% of patients who were admitted to geriatric hospitals were psychiatrically ill, only 12% of them appeared to be wrongly placed. Moreover, their outcome did not appear to be affected adversely—many patients can be cared for equally well in either type of hospital. Optimal placement depends on the relative predominance of behavioural and physical problems. Medical and psychiatric teams need to cooperate closely if all patients are to receive appropriate treatment.

Long-term hospital care. In some countries, most elderly psychiatric patients are still treated in the wards of psychiatric hospitals. However, the nature of the care provided is more important than the type of institution in which it is given. The basic requirements are opportunities for privacy and the use of personal possessions, together with occupational and social therapy. Provided that these criteria are met, long-term hospital care can be the best provision for very disabled patients. In the UK, hospital provision of long-term care has been drastically reduced (with some increase in the level of community care), and long-term care is now largely provided outside hospital in residential and nursing homes. There is an active debate on how long-term care should be funded, as care provided by the NHS is free, whereas care provided through the social services is means tested (Royal Commission on Long Term Care, 1999). A statement from the Royal Commissioners in 2003 clarified payments for nursing and personal care. However, the more definitive Dilnot Commission report is still awaited, but should be published in the lifetime of this edition of this book (Commission on Funding of Care and Support, 2011).

Social services

Domiciliary services

In addition to medical services, domiciliary services include home helps, meals at home, laundry, telephone, and emergency call systems. In the UK, local authorities both provide and commission these services; they also support voluntary organizations and encourage local initiatives such as good-neighbour schemes and self-help groups. The last decade has witnessed a major shift in provider for these services, with the majority being from private or voluntary organizations. In a random sample of nearly 500 people aged 65 years or over living at home, Foster et al. (1976) found that 12% were receiving domiciliary services but a further 20% needed them. Although these provisions are increasing, so are the numbers of those requiring them. Bergmann et al. (1978) have argued that if resources are limited, more should be directed to patients living with their families than to those living alone. This is because the former can often remain at home if they receive such help, whereas many of the latter require admission before long, even when extra help is given.

Residential and nursing care

Older people may need a variety of social service accommodation, ranging from entirely independent housing through sheltered housing schemes, where there may be some communal provision, often including access to a warden, to residential or nursing homes where there are staff available at all times. There is a need for special housing for the elderly that is conveniently sited and easy to run. Ideally, the elderly should be able to transfer to more sheltered accommodation if they become more disabled, without losing all independence or moving away from familiar places.

‘Continuing care’ refers to the provision of homes where staff are available on site at all times. In residential homes, the needs of residents for help with personal care can be met by care assistants with relatively little training, whereas in nursing homes residents will require regular nursing care, and therefore the staff will include a substantial number of trained nurses. Some, but not all, nursing homes specialize in the care of older people with mental disorders.

In the UK, local social services are responsible for providing residential homes and other sheltered accommodation. However, many independent organizations and charities also provide residential homes for older people. The role of the voluntary and charitable sector in this provision is growing, and is possibly set to become the dominant sector provider, given the direction of current national policies. The growth of continuing care provision in the private sector has encouraged a greater variety of facilities. However, undue emphasis on the cost rather than the standard of care, together with a lack of coordinated planning, has sometimes made it difficult to develop appropriately integrated local services. Over recent decades there has been a steady rise in the level of disability of residents in residential homes.

Voluntary Services

Voluntary agencies play a large and increasing role in the provision of facilities and support for patients, their families, and carers. Previously the task for NHS services was to ensure that their contributions are integrated with health and social service provisions. As their role increases and they may be set to play a dominant role in this area, the challenge facing NHS and social services may be to adapt their own services to complement, rather than just integrate, the voluntary provision. Any discussion of service provision for elderly mentally ill patients must acknowledge the major part played by informal carers.

Informal carers

Informal carers are unpaid relatives, neighbours, or friends who look after disabled (usually demented) elderly people at home. The term differentiates these people from formal carers such as paid home helps and district or community psychiatric nurses. Informal carers provide substantially more care to older people than do the statutory services, and their role in the overall provision of care should not be underestimated. Most informal carers of older people are (in descending order of frequency) partners, daughters, or sons. About twice as many women as men are informal carers, and about 50% of all informal carers are themselves elderly.

Several studies have shown that patients suffering from dementia place the greatest stress on carers (Pin-quart and Sörensen, 2003). In general, the more severe the dementia, the greater the strain experienced by the carer. Incontinence, behavioural disturbance at night, and aggression are the most distressing problems for carers, many of whom have symptoms as severe as those we would expect in a person presenting as a psychiatric case (assessed by the General Health Questionnaire). The degree of strain imposed on these elderly, committed, and often stoical individuals is often undeclared and easily underestimated. Caregiver symptoms lessen when the patient has moved to permanent residential care (Gaugler et al., 2009).

Accurate assessment of carers’ needs is important, and in the UK, social services now have a statutory responsibility to provide this, where requested. Table 18.7 summarizes the key recommendations made by Levin (1997). Time should be spent with family carers in giving advice about the care of patients and discussing their problems. Such support can help families to avoid some of the frustration and anxiety involved in caring for elderly relatives. Published guides, which are increasingly available on the Internet, are also useful. Other practical help may include day-care or holiday admissions, and laundry and meal services to the patient’s home. With such assistance many patients can remain in their own homes without imposing an unreasonable burden on their families. The assessment of need and carer support may often have to be multidisciplinary; both community psychiatric nurses and care managers (social workers) play essential roles in coordinating these services, supporting relatives, and providing direct nursing care.

Table 18.7 Support for carers of elderly patients with dementia

1. Early identification of dementia

2. Comprehensive medical and social assessment of identified cases

3. Timely referrals between agencies (e.g. from general practitioner to old age psychiatrist)

4. Continuing reviews of each patient’s needs, and back-up for carers

5. Active medical treatment for any intercurrent illness

6. The provision of information, advice, and counselling for carers

7. Regular help with household and personal care tasks

8. Regular breaks for carers (e.g. by providing day care and respite care for the patient)

9. Appropriate financial support

10. Permanent residential care when this becomes necessary

Despite the increasing focus on carers’ needs, there is little evidence that any interventions are effective in improving their psychosocial well-being (Parker et al., 2001). However, there is some support for problem-solving and behavioural management strategies (Pusey and Richards, 2001).

Psychiatric assessment in the elderly

The principles and basic purpose of assessment of older people are not substantially different from those for all other patients. Assessment is intended to establish a diagnosis, to develop the best possible plan of care, and to inform the prognosis. However, there are differences of emphasis and process.

• The assessment is likely to take longer, requiring more than one interview, and with a greater reliance on informants. For example, in the assessment of suspected dementia, an informant is essential to complete the history, and to verify the extent to which the impairments are affecting the patient’s capabilities and safety. In the evaluation of cognitive impairment, time will also be required to administer rating scales.

• The assessment is likely to extend beyond a strictly medical understanding of the patient’s condition and its treatment to include assessment of the wider psychosocial situation, taking into account not only the patient’s needs but also those of carers and other individuals who are involved. This is particularly relevant for the patient with moderate or severe dementia.

• The physical examination and laboratory investigations play a greater role, because of the higher prevalence of organic disorders as a cause of psychiatric symptoms in the elderly. The psychiatric assessment may also bring coincidental medical problems to light.

For a review of psychiatric assessment in the elderly, see Jacoby (2009).

The referral

It is important to establish at the outset what prompted the referral, and what the referrer hoped to gain from it. This may well require information to be collected even before the patient is seen, not least in order to establish the most useful way to approach the assessment. In any one case, each person involved in the patient’s care may have different, and possibly conflicting, expectations and needs.


Many old people who are seen by psychiatrists are unable to give complete or reliable information about themselves. Frequently there is a partner or other close relative living with the patient, but in other cases it may be necessary to talk to neighbours or friends. It may be useful to spend considerable time telephoning relatives or others who may be able to give information about the patient’s family history or previous personality. Many patients who are seen for the first time will already be well known to their general practitioner or to other health professionals, and it is always worth consulting them.

Where to assess the patient

In the UK, most old age psychiatrists prefer to assess patients in their own homes. This enables much essential ‘real-life’ information about the patient’s ability to function at home to be gained (e.g. ability to make a cup of tea, or recognize relatives in family photographs). It also avoids the patient appearing excessively disorientated simply because of the disturbing effect of having to travel to a hospital for assessment in an unfamiliar environment. Furthermore, it makes it easier to interview other members of the family, and to assess the level of support from neighbours or outside carers, who may be available at the patient’s home during the assessment.

Old age psychiatrists may also be asked to assess patients on general hospital wards. Although this provides less information about the patient’s circumstances, it makes close liaison with the hospital team possible. Conditions on hospital wards are often unfavourable for a quiet, private interview. It is important to spend time reading the notes carefully, talking to nursing and other staff, and then requesting the use of an office or side room in which to see the patient. Even if the family or other carers cannot be present, it may be possible to telephone them. Similar principles apply to assessment in residential and nursing homes.

Increasingly, patients with early cognitive impairments are being offered assessments in a memory clinic. This offers the advantages of a systematic and detailed assessment, and enables relevant further investigations and follow-up to be arranged as efficiently as possible. It is also the vehicle by which drugs to treat dementia are prescribed in the NHS.

The assessment

The information that is required about the medical and psychiatric history is the same as in younger patients. However, it may be necessary to piece it together from accounts given by the patient and by other informants. It is of particular importance to obtain a clear medical history and to determine the patient’s past and present medication.

During the mental state examination, assessment of cognitive function has particular significance, especially for those in whom memory impairment is apparent or suspected. This assessment requires a range of clinical questions and observations, supplemented by one or more of the questionnaires available for this purpose (see Table 13.11, p. 325), such as the Mini-Mental State Examination or the CAMCOG.

As physical problems are extremely common, the psychiatrist will often need to examine the patient thoroughly (especially if this has not been done recently by the general practitioner), including a neurological examination. Laboratory investigations are important for patients on admission, and in all others with significant psychiatric disorder (for investigation of dementia, see p. 324).

Finally, the assessment process includes assessments of risk, and of the needs of the carer(s), discussed above, as most elderly people with psychiatric disorders live in their own homes and are cared for by family members or, occasionally, by good neighbours or friends.

Psychiatric treatment in the elderly

As with assessment, the principles of psychiatric treatment in the elderly resemble those for other adults, but the greater age of these patients does require one to bear in mind three issues which have an impact upon treatment (Oppenheimer, 2009).

• Elderly patients are likely to have multiple problems. Psychiatric, physical, and social difficulties usually coexist to some extent. ‘Treatment’ may thus include a broad range of interventions beyond those normally associated with psychiatry—for example, antibiotics for a urinary tract infection, or liaison with a district nurse or dietitian. These complex treatment needs are reflected in the multidisciplinary nature of services, described above, and require careful planning and integration of treatment provision.

• Clear boundaries between normality and disease are rare. This poses challenges for treatment thresholds and service provision.

• Lack of competence is common, due to cognitive impairment. There is an increasing focus on how (lack of) competence should be assessed and responded to. The question of competence underlies a range of ethical and legal issues in which old age psychiatrists become involved (see below).

Physical treatments

The efficacy of psychotropic drugs is generally not affected by age, and elderly patients should not be denied effective drug treatment, especially for depressive disorders. However, as noted in Chapter 19, prescribing in the elderly requires particular caution, for two main reasons.

• The incidence of unpleasant or dangerous side-effects is high, and can produce delirium and other psychiatric disorders. Most problems arise with drugs used to treat cardiovascular disorders (antihypertensives, diuretics, and digoxin) and those that act on the central nervous system (antidepressants, hypnotics, anxiolytics, antipsychotics, and antiparkinsonian drugs). These problems arise because of differences in pharmacokinetics with ageing, and the greater number of drugs which elderly people are prescribed, which increases the likelihood of harmful interactions. There is particular concern about the use of antipsychotics in dementia (see p. 499).

• Compliance with treatment may be compromised in those who live alone, who have poor vision, or who are forgetful or confused.

Given these factors, the following points should be borne in mind when prescribing for the elderly person with a psychiatric disorder.

• Always start with a low dose, increase slowly, and expect the final dose to be considerably lower than in younger patients. The main exceptions to this are the SSRIs, for which comparable doses are used.

• All medication should be reviewed regularly, and kept to a minimum. Before starting a drug in the elderly, it is good practice to state clearly in the notes the reason for the decision (including, for example, the score on a depression rating scale), and the criteria by which the effects of the treatment will be assessed. If good evidence of efficacy is not achieved, medication should be gradually withdrawn.

• The drug regimen should be as simple as possible. Medicine bottles should be labelled clearly, and memory aids, such as packs containing the drugs to be taken on a single day with daily dose requirements, should be provided. If possible, drug taking should be supervised.

Electroconvulsive therapy (ECT)

ECT remains one of the most effective and valuable treatments for serious depressive disorder in the elderly. Advanced age is not a contraindication. However, attention should be paid to the physical health of all elderly patients undergoing this treatment, and frail patients should be assessed by an experienced anaesthetist before receiving ECT.

Psychological treatment

In all patients, supportive therapy with clearly defined aims can be helpful, including joint sessions with the patient’s partner or carer.

In patients who have no cognitive impairment, the specific psychological therapies that are used in younger individuals remain appropriate, for the same range of disorders and with similar expectations of success. In particular, cognitive and behavioural interventions for mood and anxiety disorders are effective, and are often preferred by the patient. Family or systems therapy, although not widely available, is advocated as it explicitly recognizes the patient in their social context (Jacoby, 2009). Psychodynamic psychotherapy is less appropriate for the elderly and is seldom used, although cognitive analytic therapy has been found useful.

Even mild cognitive impairment militates against the use of most of the common psychological therapies, as it is likely to prevent the patient from understanding or implementing the treatment (Oppenheimer, 2009). However, a range of specialized psychological and behavioural interventions (e.g. reminiscence therapy) are becoming widely used in this group, with the benefits targeted as much at the carer as at the patient.

For a review of psychological treatment in the elderly, see Wilkinson et al. (2008).

Psychosocial treatments

Some patients can achieve independence through measures designed to encourage self-care and domestic skills, and to increase social contacts. For those living at home, a domiciliary occupational therapist may be able to give useful advice on environmental or other modifications that will help the patient to live more independently. More severely impaired patients who are living in institutional care can benefit from an environment in which individual needs and dignity are respected, and each person retains some of their personal possessions. Further psychosocial interventions of this kind are discussed in the section on treatment of dementia.

Table 18.8 Ethical and legal issues in the elderly

Confidentiality in relation to information from carers

Confidentiality of information about financial


Consent to treatment

• Capacity to consent to physical and psychological treatment

• Advance directives

• Decisions ‘not to treat’

Damaging behaviour

Management of financial affairs

• Nominating another to take responsibility (Power of Attorney)

• Procedures to enable others to take responsibility

Entitlement to drive a vehicle

Consent to participate in research

Legal, financial, and ethical issues

Both the law and ethical principles apply as much to the elderly as to younger people, and these issues are discussed in Chapter 4. However, certain legal and ethical issues that are relevant in psychiatry have particular relevance to the elderly (see Table 18.8), for several reasons:

• the nature of the common illnesses from which they suffer, especially dementia, which compromises their competence to make decisions and give consent

• problems arising from increasing age and frailty (e.g. driving)

• the greater likelihood and proximity of death, which gives greater urgency to issues such as testamentary capacity (Posener and Jacoby, 2008) and advance directives.

In the UK, many of these issues are governed by the Mental Capacity Act 2005.

Financial affairs

Older people, particularly those with mental disorders, may have difficulty in managing their financial affairs. In the UK, if the issues are relatively simple—for example, involving only a state pension and benefits—the person can request that someone else, known as an appointee, collects these benefits on their behalf. Alternatively, while the person still has mental capacity to understand what is involved, they can grant a Lasting Power of Attorney in favour of one or more others. The attorney then has authority to carry out any financial transaction on behalf of that person. If the donor of the power subsequently becomes mentally unable to manage their own affairs, the attorney may continue to do so on their behalf (in the UK, provided the arrangement is registered with the Court of Protection). Sometimes a person becomes mentally incapable of managing their affairs without having made any formal arrangement for someone else to act on their behalf. Under these circumstances, an application supported by medical evidence of incapacity may be made to the Court of Protection for a deputy to be appointed to manage the patient’s affairs.

Under the Mental Capacity Act an attorney can also make proxy health and welfare decisions on behalf of an incompetent person. This requires that the person was competent to donate that power of attorney before subsequently becoming incompetent. The Mental Capacity Act has introduced a new power, the Advance Decision to Refuse Treatment (ADRT), which allows an agreed proxy to make decisions which are binding even where they will result in death. Because of the gravity of this power, it must be written and witnessed and confirm the competence of the patient at the time at which it is arranged (see Chapter 4). The ADRT can be overridden by the Mental Health Act.


A common practical problem is the inability of older patients, especially those with dementia and Parkinson’s disease, to drive safely. As with other disorders, doctors generally have an obligation, which overrides confidentiality, to inform the authorities responsible for the provision of driving licences if the patient declines to do so him- or herself. In practice, elderly people often make the decision to stop driving anyway, or are persuaded to do so by relatives. For a review, see Breen et al. (2007).

Abuse and neglect of the elderly

Abuse and neglect of the elderly by family members or other carers is an issue of increasing concern for all healthcare professionals working with the elderly. Its psychiatric relevance arises in part from the finding that people with dementia are particularly likely to be abused.

Elder abuse

This term refers to actions by a carer or other trusted person that cause harm or create a serious risk of harm to an elderly person (whether or not harm is intended), or to failure by a caregiver to satisfy the elderly person’s basic needs or protect them from harm. The term ‘elder maltreatment’ is also used.

Five forms of elder abuse are recognized—physical, psychological, sexual, financial, and neglect. Prevalence rates of 2–10% are reported, with a higher risk if:

• the abused person has dementia

• the carer and the abused person live together

• the abused person is socially isolated (e.g. lacks close friends)

• the carer has a psychiatric disorder or misuses alcohol

• the carer is heavily dependent (e.g. financially) on the person who is being abused.

Elder abuse has a threefold increased mortality rate compared with that for matched elderly people (Lachs and Pillemer, 2004), as well as with other adverse outcomes ranging from depression to placement in a nursing home.

Although there is now widespread awareness of the problem, and of the need to identify elder abuse and respond to it, there is no evidence about the effectiveness of interventions. At present, a range of measures are used, such as increased social support, respite care, relationship counselling, and carer education programmes. Legal approaches (e.g. guardianship, law enforcement agencies) may also be necessary.

For a review of elder abuse, see Hirsch and Vollhardt (2008).

Treatment of dementia

The clinical, epidemiological, and aetiological features of dementia were described in Chapter 13. Here its management will be covered. First the treatment of the behavioural and psychological disturbances of dementia, for which non-pharmacological and pharmacological strategies are used, will be discussed. Then the pharmacological treatment of Alzheimer’s disease and some other specific dementias for which drugs are becoming available will be considered.

For overviews of the management of dementia, see Taylor and Fleminger (2009), Burns and Illiffe (2009a,b), and the National Institute for Health and Clinical Excellence (2006/2011).

Treatment of behavioural and psychological symptoms of dementia

A spectrum of behavioural changes and psychological symptoms occur during the course of dementia, as described in Chapter 13 and summarized in Table 18.9. These features affect about 60% of sufferers at some stage of the illness, although the overall prevalence and frequency of individual features vary according to the type of dementia. The behavioural and psychological symptoms cause considerable problems for carers, and their management can be difficult. There is increasing concern about the inappropriate use of medication, yet there is a shortage of evidence-based non-pharmacological alternatives.

Table 18.9 Behavioural and psychological symptoms of dementia






Inappropriate sexual behaviour

Impaired sleep






Contributory factors



Superimposed delirium

Sensory deficits

Before any treatment is considered, careful assessment is required, and attention should be paid to the following points:

• The nature of the problem behaviour(s) should be clearly identified, including its duration, severity, and any suspected causative or modifying factors. Rating scales may be useful (see Table 13.11, p. 325). Examples of remediable causal factors include urinary tract infections, constipation, and pain. The latter is often overlooked and adequate analgesia therefore not provided (Scherder et al., 2005). Pain in dementia may be due to concurrent physical problems (e.g. leg ulcers, arthritis), or it may be part of the disease process.

• A mental state examination is needed to determine whether the behaviour can be explained in terms of an underlying psychopathology (e.g. whether there is evidence of psychotic symptoms, low mood, or anxiety). If this examination, together with the history and other observations, supports a diagnosis (e.g. of psychosis, depression, or delirium), it will probably influence the treatment plan. On other occasions, however, no such diagnostic evidence may be forthcoming, in which case the behaviour is dealt with on its merits.

• The natural history is that behavioural and psychological symptoms of dementia tend to fluctuate and often last for less than 3 months (Hope et al., 1999). Any treatment, especially if drugs are used, should reflect this fact. Regular reassessment, and trials of medication withdrawal, should be an integral part of the management plan.

• Behavioural changes in dementia are usually problems for the carer rather than for the patient. This reality should be borne in mind when treatment decisions are made, especially as informed consent is often not possible because of the patient’s cognitive impairment.

Non-pharmacological treatment of behavioural and psychological symptoms

If a behaviour or symptom is deemed, following assessment, to be sufficient to require treatment, a hierarchical approach should be employed. The first approach is to treat any underlying remediable cause, as outlined above. If the problem persists, the next approach should be based on modification of the patient’s environment, or a behavioural or psychological intervention, the choice of which will be influenced by the nature of the problem, the severity of the dementia, the setting, and the resources and expertise available (see Table 18.10). Medication should be reserved for those patients in whom such approaches prove unsuccessful.

Most of the interventions listed in Table 18.9 have some evidence of effectiveness. However, there is little substantial evidence from randomized controlled trials. One study showed that an intervention consisting of exercise and a carer training programme improved depression and physical health in 153 patients with Alzheimer’s disease, with a trend towards less institutionalization for behavioural problems 2 years later (Teri et al., 2003). More recently, a multifaceted ‘biobehavioural’ intervention was shown to improve the well-being of patients with dementia and their carers (Gitlin et al., 2010).

For reviews of non-pharmacological interventions in dementia, see Livingston et al. (2005) and James and Fossey (2008). For a review of strategies for use in primary care, see Robinson et al. (2010).

Table 18.10 Non-pharmacological interventions for behavioural and psychological symptoms of dementia

Sensory stimulation

Music therapy

Aromatherapy (e.g. lavender oil)

Bright light therapy


White noise

Behavioural management

Differential reinforcement

Stimulus control

Social contact

One-to-one interaction

Pet therapy

Simulated contact (e.g. videos)


Structured activity programmes

Environmental modifications

Wandering areas

Carer education and support

Combination therapies

Drug treatment of behavioural and psychological symptoms

High rates of prescribing of psychotropic drugs in dementia, largely to control behavioural problems, have been reported in nursing homes and other institutions. For example, a study conducted in 1996 found that up to 30% of residents in Glasgow nursing homes were receiving antipsychotics (McGrath and Jackson, 1996), as were 18% of those with dementia in a large American survey (Liperoti et al., 2003). However, the evidence shows that medication has at most a modest effect and, in the case of antipsychotics, is associated with considerable morbidity and an excess mortality, as noted below. Reflecting these increasing concerns, medication for behavioural problems and psychological symptoms in dementia should not be contemplated until:

• a full assessment has been conducted

• any aggravating physical and environmental factors have been addressed

• non-pharmacological strategies have been attempted. In a randomized trial, Fossey et al. (2006) showed that enhanced psychosocial care led to a significant reduction in the prescribing of medication for nursing home residents with dementia.

Table 18.11 summarizes the drug treatments that are currently used for behavioural problems in dementia (Burns and Illiffe, 2009a; see also Ballard et al., 2009a).


Despite their widespread use, robust evidence for the efficacy of antipsychotics in the behavioural problems associated with dementia is limited to the short-term (6–12 weeks) treatment of aggression (Ballard et al., 2009a). These limited benefits must be set against the significant risk of extrapyramidal side-effects, postural hypotension, tardive dyskinesia, metabolic syndrome, and the many other side-effects of antipsychotics to which the elderly are prone (Schneider et al., 2006). The increasing concern about the use of antipsychotics in dementia is enhanced by three other serious potential risks.

1. They may increase the risk of stroke. It remains unclear whether atypical antipsychotics are associated with a higher risk than typical antipsychotics in this regard (Gill et al., 2005; Mazzucco et al., 2008).

2. Antipsychotics can precipitate an irreversible and sometimes fatal syndrome of Parkinsonism, impaired consciousness, and autonomic disturbance, notably in patients with dementia with Lewy bodies (McKeith et al., 2004). This effect, together with the stroke risk, may underlie the finding that antipsychotic prescribing increases the risk of death in patients with dementia (Schneider et al., 2005; Ballard et al., 2009b).

3. Antipsychotics may hasten cognitive decline (McShane et al., 1997), an observation that has been confirmed in a randomized trial of quetiapine (Ballard et al., 2005). The mechanism involved is unknown, but may relate to anti-muscarinic actions.

Reflecting these concerns, current UK guidelines recommend that antipsychotics should not be used in dementia unless the patient’s problems are severe, associated with either psychotic symptoms or serious distress, or the behaviour poses a danger of physical harm. Comparable warnings and restrictions are also in force elsewhere, and have led to a fall in prescribing rates (Kales et al., 2011). However, many elderly people, especially those in nursing homes, continue to be prescribed these drugs (Shah et al., 2011). Antipsychotics should be used rarely, if ever, in dementia with Lewy bodies. Assessment for vascular risk factors is advised because of the risk of stroke. As elderly people are more sensitive to antipsychotics, treatment should always start with a low dose, which is increased slowly, and the effects should be monitored regularly (Uchida et al., 2009). The choice of antipsychotic should be based on the side-effect risk profile, with particular attention to extrapyramidal and anticholinergic effects and sedation.

Cholinesterase inhibitors and memantine

These drugs were developed to treat the cognitive symptoms of Alzheimer’s disease. They are used to treat behavioural problems and psychological symptoms of dementia, too, in part because of the increasing concern about antipsychotics. However, it is unclear whether cholinest-erase inhibitors are effective against these aspects of dementia. A recent meta-analysis was inconclusive (Rodda et al., 2009), although they may have a role in behavioural disturbances occurring in dementia with Lewy bodies. The evidence with regard to memantine in the treatment of behavioural and psychological symptoms is rather more convincing (Wilcock et al., 2008).

Other drugs

A range of other drugs are also used, as illustrated in Table 18.11, based upon limited evidence (Ballard et al., 2009b). For depression in dementia, antidepressants show equivocal efficacy (Modrego, 2010).

Drug treatment of Alzheimer’s disease

Treatment of behavioural and cognitive symptoms continues to be the most important element of therapy of dementia. However, pharmacological treatment of Alzheimer’s disease itself is now available with the advent of cholinesterase inhibitors and memantine, and a range of other drugs are under investigation.

Table 18.11 Daily doses of drugs used to treat behavioural and psychological symptoms of dementia

For mild agitation

Trazodone 50–100 mg

Benzodiazepines, e.g. lorazepam 0.5–4 mg

SSRIs, e.g. citalopram 10–20 mg

Carbamazepine 50–300 mg

Sodium valproate 250–1000 mg

Rivastigmine (for Lewy body dementia) 1.5–6 mg

For severe agitation with psychosis

Quetiapine 25–200 mg

Risperidone 0.5–3 mg

Olanzapine 2.5–10 mg

For depression

SSRI, as above

Mirtazapine 15–45 mg

For severe behavioural problems

Consider haloperidol in small doses (0.5–4 mg), for a limited time

From Burns and Illiffe (2009a).

For clinical guidelines about the drug treatment of Alzheimer’s disease, see O’Brien and Burns, on behalf of the BAP Dementia Consensus Group (2011). For guidelines on the broader principles of management of dementia, and the context in which drugs are used, see the National Institute for Health and Clinical Excellence (2006/2011).

Cholinesterase inhibitors

A loss of acetylcholine was the first neurochemical abnormality of Alzheimer’s disease to be discovered (see p. 330), and led to the development of cholinesterase inhibitors as the first effective treatments for the disease (Francis et al., 1999). They now have an important role in clinical practice, and their discovery was responsible for bringing a new therapeutic optimism to old age psychiatry. They work by increasing the survival and thus the availability of acetylcholine at the synapse; some may also enhance activity at nicotinic–cholinergic receptors. However, their efficacy is limited, they do not alter the disease process, and the magnitude of their clinical benefit, as well as their cost-effectiveness, have been controversial, reflected in the UK by shifts in their availability within the NHS (see below).

The first cholinesterase inhibitor was tacrine, but its use was limited by hepatotoxicity. The drugs currently available are donepezil, rivastigmine, and galantamine. All of these have received regulatory approval, for use in mild to moderate Alzheimer’s disease, based upon trial data showing beneficial effects, relative to placebo, on cognitive function (usually measured on the cognitive portion of the Alzheimer’s Disease Assessment Scale, ADAS-Cog) and on global functioning (assessed with the Clinician Interview-Based Impression of Change scale with carer input, CIBIC-Plus). The size of the effect is relatively modest—for example, benefits of about 3 points on the ADAS-Cog (range of scores, 0–70), and 0.4 on the CIBIC-Plus (range of scores, 1–7). As disease progression averages about 7 points on the ADAS-Cog per year, cholinesterase inhibitors may be viewed as delaying progression by about 6 months. In other words, the drugs roughly double the likelihood of a patient improving by 4 ADAS-Cog points. The efficacy of the three drugs is similar, but there are no head-to-head comparisons.

Given the inexorable progression of dementia, and the limitations of rating scales alone, an alternative and attractive way to view medication efficacy is in terms of whether it reduces the percentage of patients who ‘clinically deteriorate.’ Clinical deterioration can be defined as cognitive decline, plus a decline in activities of daily living, plus a decline in global functioning. Using these criteria, donepezil significantly reduces clinical worsening (from 30% on placebo to 14% on donepezil) over a 24-week period, with a greater proportional benefit among patients with milder dementia (MMSE > 18, from 21% to 7%) (Wilkinson et al., 2009).

Side-effects of cholinesterase inhibitors are relatively common and can be troublesome. They include nausea, vomiting, diarrhoea, muscle cramps, insomnia, bradycardia, syncope, and fatigue. Their occurrence can be reduced by starting with a low dose, taking the drug with meals, and increasing the dose slowly. If a patient does not respond to, or cannot tolerate, a cholinesterase inhibitor, a trial of a different one is worthwhile (O’Brien and Burns, on behalf of the BAP Dementia Consensus Group, 2011).

Current UK practice, reflecting guidelines from the National Institute for Health and Clinical Excellence (NICE), is that treatment with a cholinesterase inhibitor should be instituted in memory clinics by a specialist (usually an old age psychiatrist, or a neurologist or geriatrician), following a formal assessment of the patient, from which the diagnosis of Alzheimer’s disease is made, and its severity established, using the MMSE and other cognitive scales. Since March 2011 it has been possible to prescribe cholinesterase inhibitors within the NHS under these circumstances to patients who have been diagnosed with mild or moderate Alzheimer’s disease. The response should be assessed 2 to 4 months later, and the drug continued only if the MMSE score has not declined, together with evidence of global or behavioural improvement. In those who show this initial response, the drug should continue as long as the MMSE score remains above 12, taking into account the patient’s overall condition and global functioning. Decisions to stop medication are complicated by the possibility that this is followed by an accelerated cognitive decline, such that within months the patient returns to the expected level of functioning had treatment never been given.

Cholinesterase inhibitors are not recommended for treatment of mild cognitive impairment (see p. 500).


Memantine is an N-methyl-D-aspartate (NMDA)-type glutamate-receptor antagonist. The rationale for its use in Alzheimer’s disease is that neurotoxicity mediated via the NMDA receptor might contribute to the disease process (Parsons et al., 2007).

Randomized controlled trials have shown that memantine benefits cognition and function in moderate to severe Alzheimer’s disease (McShane et al., 2006; Winblad et al., 2007), in addition to its positive effects on psychological and behavioural symptoms mentioned earlier. Applying the concept of clinical worsening mentioned with regard to cholinesterase inhibitors, memantine significantly reduces the risk compared with placebo (11% vs. 21%) (Wilkinson and Andersen, 2007). Combining memantine with a cholinesterase inhibitor may provide additional benefits, although this has yet to be established (Lopez et al., 2009; O’Brien and Burns, on behalf of the BAP Dementia Consensus Group, 2011).

In the NHS, memantine can now be prescribed for patients with severe Alzheimer’s disease, and for those with moderate Alzheimer’s disease who cannot tolerate cholinesterase inhibitors.

Other drugs

As discussed in Chapter 13, exposure to hormone replacement therapy, non-steroidal anti-inflammatory drugs, and statins have all been associated with lower rates of Alzheimer’s disease (see p. 329). Each of these drugs has therefore been used to treat the condition, but without benefit. Vitamin E has also not been demonstrated to be effective. None of these drugs are recommended for clinical practice (O’Brien and Burns, on behalf of the BAP Dementia Consensus Group, 2011).

Extracts from the leaves of the maidenhair tree, Ginkgo biloba, have long been used in China as a remedy for various diseases. Despite enthusiasm about these extracts, it remains unclear from clinical trials and meta-analyses whether they have efficacy in Alzheimer’s disease, although they are well tolerated (Ballard et al., 2011).

Novel treatment approaches

Treatments currently under development are intended to modify the disease process, and thereby produce greater and more persistent benefits than the drugs that are being used at present.

The increasing evidence that β-amyloid is central to the Alzheimer’s disease process (see p. 329) has led to this molecule and its biochemical pathways becoming the key treatment targets. Three main approaches are being used:

• drugs to inhibit the enzymes (secretases) by which β-amyloid is formed from its precursor molecule

• drugs to prevent aggregation of β-amyloid

• immunization strategies that target β-amyloid. This approach arose from unexpected findings in a mouse model of the disease, discussed in Box 13.1 (see p. 328).

The other main therapeutic target is tau protein and its phosphorylation (see p. 328). To date, none of these strategies have produced clear positive results in clinical trials, although a large number are under way.

For reviews of current drug treatment research in Alzheimer’s disease, see Mangialasche et al. (2010) and Ballard et al. (2011).

Drug treatment of other dementias

There are no drugs currently licensed for any other common dementia. However, several of the approaches that are being used in Alzheimer’s disease are also being applied in these disorders, and some are mentioned in dementia treatment guidelines (National Institute for Health and Clinical Excellence, 2006/2011; O’Brien and Burns, on behalf of the BAP Dementia Consensus Group, 2011).

Vascular dementia

There have been few randomized trials with regard to the treatment of vascular dementia. Cholinesterase inhibitors and memantine may produce small cognitive improvements but no benefit in global outcome in vascular dementia, and they are not recommended for use (O’Brien and Burns, on behalf of the BAP Dementia Consensus Group, 2011). Similarly, there are insufficient data to support use of the calcium-channel blocker nimodipine, or huperzine A (a natural cholinesterase inhibitor derived from a Chinese herb). Currently, treatment of vascular dementia remains limited to control of the cardiovascular risk factors that are thought to underlie the condition. For a review, see Kirshner (2009).

Dementia with Lewy bodies

Cholinesterase inhibitors and memantine appear to have similar effectiveness in dementia with Lewy bodies and in Parkinson’s disease dementia to that which they have in Alzheimer’s disease. The concurrent beneficial effects of cholinesterase inhibitors on the behavioural symptoms of dementia with Lewy bodies mentioned earlier are advantageous, given the toxicity of antipsychotic drugs in this patient group. For a review, see Kurtz and Kaufer (2011).

Other dementias

In frontotemporal dementia, no treatments have been shown to improve cognition (or behavioural symptoms). Cholinesterase inhibitors are not effective and may worsen the behavioural symptoms. No treatments are known for prion disease (a trial with quinacrine proved negative), for progressive supranuclear palsy, or for the cognitive component of Huntington’s disease, although a range of drug interventions for its psychiatric and motor components are available (Phillips et al., 2008).

Finally, despite much interest in delaying the onset of dementia, no treatment is yet known or recommended for mild cognitive impairment. Cholinesterase inhibitors and vitamin E are not effective.

Clinical features and treatment of psychiatric disorders in the elderly

Depressive disorder in the elderly

Clinical features

There is no clear distinction between the clinical features of depressive disorders in the elderly and those in younger people, but some symptoms are more striking in the elderly. Post (1972) reported that one-third of depressed elderly patients had severe retardation and agitation. Depressive delusions concerning poverty and physical illness are common, and occasionally there are nihilistic delusions, such as beliefs that the body is empty, non-existent, or not functioning (see Cotard’s syndrome, p. 208). Hallucinations of an accusing or obscene kind may occur. Depression itself is sometimes not conspicuous and may be masked by other symptoms, particularly hypochondriacal complaints. Depressive disorder in the elderly should always be considered when the patient presents with anxiety, hypochondriasis, or confusion. For a review, see Baldwin (2008).

Depressive pseudodementia

Some retarded depressed patients present with ‘pseudo-dementia’—that is, they have conspicuous difficulty with concentration and remembering, but careful clinical testing shows that there is no major defect of memory. Differentiation of these patients from those with early dementia (in whom depression may occur) is important and can be difficult. Moreover, the distinction is complicated by evidence that depressive pseudodementia is a strong predictor of subsequent dementia (Saez-Fonseca et al., 2007). Features that suggest depressive pseudodementia include the following:

• the patient’s complaint of memory disturbance tends to be greater than the informant’s account of memory problems in everyday life

• depressive symptoms that pre-dated the memory difficulties.

• ‘Don’t know’ responses and poor involvement with neuropsychological tests (these features are characteristic)

• a personal or family history of mood disorder.


The aetiology of depressive disorders of first onset in late life broadly resembles the aetiology of similar disorders at younger ages, with biological (Krishnan, 2002), psychological, and social influences (Blazer and Hybels, 2005). Twin studies show a moderate heritability, which is comparable to but somewhat lower than that for depression of early onset, suggesting a slightly smaller genetic contribution (Johnson et al., 2002).

It might be expected that the loneliness and hardship of old age would be important predisposing factors for depressive disorder. Surprisingly, there is no convincing evidence for such an association (Murphy, 1982). Indeed, Parkes et al. (1969) even found that the association between bereavement and mental illness no longer held in the aged.

A role for vascular factors in depression in the elderly was noted by Post (1972), and has become the prominent hypothesis since Alexopoulos et al. (1997) coined the term ‘vascular depression.’ There is now strong and diverse evidence for a multifaceted and bidirectional relationship between late-life depression and vascular disease and its risk factors (Thomas et al., 2004). This evidence is both epidemiological and biological. The latter includes findings that the neuropathological basis of the white-matter hyperintensities which are frequently seen in late-life depression on MRI scans reflects focal areas of ischaemia and infarction. Some evidence supports the view that ‘vascular depression’ is a discrete subgroup, rather than vascular factors simply being part of the biological contribution to depression in the elderly in general (Sneed et al., 2008). The presence of vascular risk factors appears to be associated with a poorer treatment response, more cognitive impairment, and poorer outcome of late-life depression (Baldwin, 2008).

Differential diagnosis

As noted above, it is sometimes difficult to distinguish between depressive pseudodementia and dementia, and this problem is compounded by the fact that depression may occur during the course of dementia. It is essential to obtain a detailed history from other informants and to make careful observations of the patient’s mental state and behaviour using the differentiating features listed on p. 503. Psychological testing can be a useful adjunct, but it requires experienced interpretation and it usually adds little to skilful clinical assessment. At times, dementia and depressive illness coexist. If there is real doubt, a trial of antidepressant treatment may be appropriate. Other aspects of the differential diagnosis are similar to those for depressive disorder in younger people (see Chapter 10).


The principles of the treatment of depressive disorder are the same as for younger adults, as described in Chapter 10. Effective treatments thus exist, but often they are not implemented, both because cases are not diagnosed and because the interventions are not provided (Chew-Graham et al., 2004). Guidelines on the management of late-life depression in primary care have been produced (Baldwin et al., 2003). With elderly patients it is especially important to be aware of the risk of suicide. Any intercurrent physical disorder should be thoroughly treated.

Antidepressants appear to be as effective as in younger people (Katona and Livingston, 2002). The drugs should be used cautiously, perhaps starting with half the normal dosage and increasing this slowly in relation to side-effects and response. SSRIs are preferred to tricyclic antidepressants because of their lower propensity to cause side-effects and cardiotoxicity. Potential interactions with other drugs that are being taken by the patient must also be noted, and may influence the choice of antidepressant (Spina and Scordo, 2002). For patients who do not respond to the maximum tolerated dose of an antidepressant, it may be necessary to use a combination of drugs, as in younger patients, or to combine medication with psychological treatment.

As with medication, psychological interventions show similar efficacy in the elderly, and are often preferred by patients, but their availability is very limited in primary care, where most cases are managed. There is evidence that, within this setting, the outcome is improved by models of collaborative care in which specialists provide some input, and care is supplemented by education about depression and medication (Unützer et al., 2002).

ECT is appropriate for depressive disorder with severe and distressing agitation, suicidal ideas and behaviour, life-threatening stupor, or failure to respond to drugs. Concerns that ECT is dangerous or ineffective in elderly patients are unwarranted, and it should be used with a similar threshold to that in younger patients, albeit sometimes at a lower frequency.

After recovery, expert guidelines suggest that full-dosage antidepressant medication should be continued for at least 12 months for a first episode in the elderly, and proportionately longer for subsequent episodes. Indefinite continuation has also been advocated.

For a review of the management of depression in the elderly, see Baldwin (2008).

Course and prognosis

In a classic study, Post (1972) described depression in the elderly as following the rule of threes—a third recover, a third remain the same, and a third deteriorate and become ‘chronic invalids.’ Later studies have generally found a similar, if not poorer, outcome (Cole et al., 1999). For example, in a 6-year longitudinal study of 277 elderly depressed people, Beekman et al. (2002) found that 32% had a severe, chronic course, 44% had a fluctuating course, and only 23% achieved substantial remissions. Even the ‘sub-threshold’ group, noted earlier, had a poor outcome. The natural history of depression in old age has implications for its management (see below).

Factors that are believed to predict a good prognosis for depression in old age include the following:

• onset before the age of 70 years

• short duration of illness

• good previous adjustment

• the absence of disabling physical illness

• good recovery from previous episodes

• religiosity (self-reported faith, or belonging to a religious group).

Poor outcome is associated with the severity of the initial illness, the presence of organic cerebral pathology, poor compliance with treatment, and severe life events during the follow-up period. As discussed in Chapter 16, the risk of suicide remains high in elderly people with depression, especially men.

Depression and dementia. There is no evidence that late-life depression is a prodrome of dementia (Schweitzer et al., 2002), notwithstanding the finding mentioned earlier that depressive pseudodementia increases the risk of dementia several years later. There is also increasing evidence that a history of depression (at any age) doubles the risk of developing dementia, notably Alzheimer’s disease (Ownby et al., 2006).

Bipolar disorder in the elderly

Broadhead and Jacoby (1990), in the first prospective study of mania in old age, found that the clinical picture was the same as in younger patients, but that a depressive episode immediately after the manic episode occurs more frequently. Mixed affective episodes may be somewhat more common than in younger patients, and organic factors play a greater role (Depp and Jeste, 2004). Management of bipolar disorder in the elderly follows the principles described in Chapter 10. Lithium prophylaxis is valuable, but the blood levels should be monitored with particular care and kept at the lower end of the therapeutic range (0.4–0.6 mmol/l). For a review, see Shulman and Herrmann (2008).

Anxiety disorders in the elderly

In later life, anxiety disorders are seldom causes for referral to a psychiatrist, although this may be largely because of non-presentation by patients and lack of recognition or referral by general practitioners. It does not reflect their rarity in the population, as shown by the epidemio-logical data reviewed earlier in the chapter.

Symptoms of anxiety disorders among the elderly are often non-specific, with features of both anxiety and depression. Hypochondriacal symptoms may be prominent (Lindesay, 2000). As in younger patients, the possibility that anxiety symptoms are secondary to an underlying depressive episode should always be actively considered before the diagnosis is made.

Personality disorder is a predisposing factor, while physical illness and life events such as retirement or a change of accommodation may act as precipitants. A first onset of panic attacks in an elderly person should always prompt a search for an underlying physical or depressive disorder (Flint and Gagnon, 2003).

The approach to treatment is similar to that for anxiety disorders in younger adults (Alwahhabi, 2003; Flint and Gagnon, 2003), with a preference for psychological and behavioural interventions, and with cautious use of medication.

Schizophrenia-like disorders in the elderly

Schizophrenia usually begins in early adulthood (see Chapter 11), but new-onset cases after the age of 60 years, although uncommon, are well recognized. This section will focus on the features that characterize such cases compared with schizophrenia of younger onset. It also highlights some issues that affect the management of all patients with schizophrenia as they become elderly (sometimes called ‘graduates’). The complicated terminology in this area has been covered earlier in this chapter (and see Box 12.2), and reflects the long-standing view that these cases are, to some extent, clinically different from earlier-onset cases.

For reviews of late-onset schizophrenia, see Tune and Salzman (2003) and Howard (2008).

Clinical features

Although the core symptoms and overall clinical profile are very similar to those of younger-onset cases, the relative prominence of some features does differ (see Table 18.12). Of these, the extreme rarity of thought disorder and negative symptoms is the most striking. A similar generalized cognitive impairment is seen regardless of age of onset, but late-onset cases show somewhat milder deficits, especially in working and verbal memory. This profile contrasts with the marked decline in intellectual performance and development of moderate to severe dementia which occurs in a significant minority of elderly patients whose illness began earlier in life (Harvey, 2001; see p. 259).

Course and prognosis

There are no contemporary data on the course or outcome of schizophrenia with onset in old age. There is no good evidence that it heralds or progresses to dementia.

Differential diagnosis

Compared with younger-onset schizophrenia, there is a greater likelihood that an elderly patient presenting with psychotic symptoms has an organic psychosis of some kind. Therefore all elderly patients who present with a first onset of schizophrenia-like features should be assessed carefully in order to exclude a delirium, dementia, or organic psychosis due to a neurological or medical disorder such as neurosyphilis, HIV, or neoplasm (see Chapter 13). As in younger patients, schizophrenia must also be distinguished from other psychiatric disorders in which psychotic symptoms occur, notably delusional disorder or affective psychosis. Such distinctions may be particularly difficult if the patient presents with persecutory beliefs but no other clear symptoms, and if there is a history suggestive of paranoid personality traits. For those in whom visual hallucinations are an isolated symptom, consider Charles Bonnet syndrome (see p. 506).

Table 18.12 Features characteristic of late-onset compared with early-onset schizophrenia

Symptoms which are more common

Visual, tactile, and olfactory hallucinations

Third-person, running commentary, and derogatory auditory hallucinations

Persecutory delusions

Partition delusions

Symptoms which are less common

Formal thought disorder

Affective flattening and blunting

Negative symptoms

Other features

Female predominance

No clear genetic loading

Association with sensory deficits and social isolation

Less premorbid educational and psychosocial impairment

Less working memory and verbal learning impairment

Much lower antipsychotic doses required

Adapted from Howard et al. (2000).


Familial aggregation is much less common than in earlier-onset schizophrenia, which suggests that there is a smaller genetic predisposition. There is no familial association with neurodegenerative or cerebrovascular disorders. Brain imaging changes are similar in nature and magnitude to those described in younger-onset schizophrenia. Unlike late-life depression, rates of white-matter hyperintensities are not increased. A role for sex hormones in the aetiology of late-onset schizophrenia has been proposed (Seeman, 1997), based upon the female predominance (which approaches 3:1), but there is little direct evidence for this. Schizoid or paranoid personality traits are common in elderly people who develop late-onset psychotic disorders (Kay et al., 1976). Deafness is also a risk factor.


Antipsychotic medication is the mainstay of treatment of late-onset schizophrenia, as in younger patients. However, much lower doses (10–20% of a ‘normal’ dose) are often sufficient. This is not just a reflection of the age of the patient, as higher doses are typically needed for earlier-onset cases when they reach the same age. Another reason for cautious use of antipsychotic medication is that the risk of tardive dyskinesia appears to be considerably greater than in younger patients (Jeste et al., 1995), although this may be independent of age at onset of illness. Atypical antipsychotics probably have a lower risk of tardive dyskinesia than do typical antipsychotics in this age group, but recent concerns that they may increase cerebrovascular events in the elderly must be taken into account (see p. 499). Cognitive and other psychological interventions may reduce disability, as in earlier-onset cases, although there is little direct evidence for this (Schimming and Harvey, 2004).

Management of schizophrenia in the elderly must also focus due attention on the complex medical and social needs of many patients; services are often fragmentary and inadequate, in part because of the emphasis on younger patients (McCleery, 2008).

Personality disorder in the elderly

As discussed in Chapter 7, some personality traits and disorders attenuate with age, while others remain or become exaggerated. A meta-analysis found that the prevalence of personality disorder in people over 50 years of age was 7–10% (Abrams and Horowitz, 1996), somewhat lower than in younger adults. The decline is mainly attributable to a decline in cluster B personality disorders (Cohen et al., 1994), whereas obsessive–compulsive and schizoid characteristics may become more prominent (Engels et al., 2003). Schizoid or paranoid traits may become accentuated by the social isolation of old age, sometimes to the extent of being mistaken for a delusional disorder or schizophrenia. Criminal behaviour is unusual, with only 1.7% of men in England and Wales who are found guilty of indictable offences being aged 60 years or older (Fazel et al., 2001).

For a review of personality disorder in the elderly, see Holroyd (2009).

Other psychiatric syndromes of the elderly

Senile squalor syndrome

Senile squalor syndrome, also known as Diogenes’ syndrome (Clark et al., 1975), is characterized by severe neglect of self and surroundings, domestic squalor, and social withdrawal and isolation. Syllogomania (the hoarding of rubbish) also occurs. It may be precipitated by stressful life events, and is associated with a high mortality after hospital admission. Other causes of self-neglect must be excluded, notably dementia, psychosis, or severe depression. The person usually stubbornly refuses all offers of help, and it is difficult to decide when to intervene using compulsory powers.

For a review, see Pavlou and Lachs (2006).

Charles Bonnet syndrome

Charles Bonnet syndrome refers to the occurrence of visual hallucinations without any other features of psychosis, dementia, or delirium. The syndrome is important, as misdiagnoses, especially of psychosis, resulting in inappropriate treatments have been reported. It is named after a Swiss philosopher who described the case of his own grandfather in 1760.

The syndrome is particularly common in elderly people with failing eyesight, in whom there is a prevalence of 10–15%. It is usually thought to involve direct damage to the visual system. It is said to occur in up to 3.5% of referrals to old age psychiatrists.

The visual hallucinations are well formed, elaborate, vivid, and perceived in external space. They may be transient or persistent, and either variable in content or stereotyped. They have no personal meaning for the patient. Although they meet most of the criteria for true hallucinations, the patient usually retains some insight into their unreality, and can often make the image disappear by closing their eyes.

The condition tends to improve if vision is restored, or when total blindness occurs. There is no specific treatment. It has been suggested that the syndrome may be an early indication of dementia, but this is unproven.

For a review, see Jacob et al. (2004).

Further reading

Gelder MG, Andreasen NC, Lopez-Ibor JJ Jr and Geddes JR (eds) (2009). Section 8: The psychiatry of old age. In: The New Oxford Textbook of Psychiatry. Oxford University Press, Oxford.

Jacoby R, Oppenheimer C, Dening T and Thomas A (2008). The Oxford Textbook of Old Age Psychiatry. Oxford University Press, Oxford. (The definitive textbook of this subject, from which many of this chapter’s citations are drawn.)

Ames D, Burns A and O’Brien J (2011). Dementia, 4th edn. Hodder Arnold, London.