Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

Behaviour problems and assessment in dementia

  • Comment

VOL: 98, ISSUE: 47, PAGE NO: 32

Graham A. Jackson, MB, ChB, MRCPsych, MRCGP, is consultant old age psychiatrist, Leverndale Hospital, Glasgow

As dementia progresses, insight may be lost, so that the need for help cannot be appreciated. Attempts to provide help may cause frustration and anger both for those providing care and for the person affected. Part 3 of this series looks at typical behavioural problems associated with dementia, methods of assessment and types of medication used. The issue of consent to treatment is also discussed.

Behaviour problems

People with dementia may develop many behavioural problems. The causes are often multiple and include a combination of loss of abilities, loss of insight and lack of inhibition. Inappropriate care, whether in terms of the environment or in terms of the ability of the carers, may compound, or in many cases even cause, the problems. The term ‘challenging behaviour’ is often used, as the problem may be that the person’s behaviour is ‘challenging’ the ability to provide care. Some of the more common behaviour problems are listed in Box 1.

Antisocial behaviour and its management

Inappropriate sexual behaviour and toilet habits and aggression can be difficult to manage in any care setting. The development of these behaviour problems is often the stage at which people with dementia first present to the secondary care services.

However, the new drugs available today mean that patients and their carers tend to engage with services at an earlier stage then previously. Early contact with services is much preferable, as it means that by the time behaviour problems occur, a good working relationship has been established with the primary care team. Future problems may also have been predicted and thus become more easily managed.

The person with dementia often moves along a hierarchy of care settings, starting with living alone, to living with relatives, to residential care, nursing home care and finally to specialised, perhaps hospital-based, care. The presence of behaviour problems is the main determinant of such moves.

Dementia is usually a slow, progressive condition, therefore sudden changes in behaviour or in cognitive abilities are likely to be caused by other problems. These include infection, other illness or even discomfort from constipation or arthritic pain. The person may have difficulty explaining this, therefore healthcare professionals must look out for the possibility that these problems are the cause of the changed behaviour, and address them accordingly.

One question that should be asked by a health professional when assessing a person with dementia is: who has the problem - the patient or the carer(s)? Although restlessness, agitation or aggression may sometimes be the result of distress or discomfort, more often this is not the case and the problem is really the effect the patient is having on others. Thus, although he/she may quite happily be displaying disinhibition, urinating in the wrong place, or shouting and swearing, such activities present difficulties for the carers or for other residents of a home.

So who has the problem? This becomes a particular issue when medication is being considered. There can be few situations in medicine where a doctor might be asked to prescribe something to someone for a condition that affects others rather than the intended recipient - and where the person is often unable to give informed consent.

Patient dignity must be considered when attempting to reduce antisocial behaviour. Patients who strip off their clothes or masturbate in a communal area may not be upset by their actions, but it may alter others’ perceptions of them. How, then, should such difficult behaviour be managed? The first and most important principle in the management of behaviour problems is to decide whether the action is a problem. It may occur because of a lack of comprehension or tolerance on the part of others.

Risk assessment is an issue. Should someone be allowed to wander the streets, use a cooker or go downstairs unaccompanied when it is known that there are risks to allowing that person to do so? There is a line between allowing freedom, choice and dignity, and danger, but who can confidently say when any level of risk becomes unacceptable?

There are many ways of approaching difficult behaviour. Too often, medication is seen as a first resort. However, appropriate use of medication may be necessary. Some aspects of this are discussed below.


Changes in carers, in the environment, in tolerance, in expectations and in many other spheres may lead to behavioural problems becoming less of an issue. However there are situations where this fails, and when the use of medication is unavoidable. There are certain principles that must be followed:

- Start low, go slow - low doses should be used first, and time given for any effect to occur

- Be aware of side-effects

- Ask whether the person is able to give consent

- Use the drug for as short a time as possible.

Most behaviour is episodic, that is, people are not usually constantly aggressive, only on occasions, and such episodes are usually short. Medication, therefore, should be prescribed only after a great deal of thought. If outbursts are common, perhaps medication could be used to lower the threshold for agitation.

In Scotland, the Scottish Intercollegiate Guidelines Network group (SIGN, 1998) looked at the evidence for efficacy of drugs (and for other interventions) and concluded that there was little evidence on which to base recommendations (Box 2). Neuroleptic drugs such as thioridazine (now accepted to be inappropriate because of its side-effects), chlorpromazine and haloperidol have been the traditional drugs used, in spite of little supporting research evidence.

One major problem with these drugs is their side-effects, particularly with prolonged use; for example, tardive dyskinesia (involuntary, repetitious movements of the muscles of the face, the limbs and the trunk), akathisia (presenting as restlessness), and Parkinsonism (a neurologic disorder characterised by tremor, muscle rigidity, hypokinesia and a slow, shuffling gait).

The newer atypical antipsychotic drugs include risperidone, olanzapine and quetiapine, but these, too, are not without problems when used in older people. Sedatives such as diazepam, lorazepam and chlor-methiazole are also used, and antidepressants such as trazodone and citalopram may be useful. Use of the mood stabilisers carbamazepine, sodium valproate and lithium has been reported in smallscale studies but again robust evidence of benefit is lacking.

Prescribing drugs for patients with dementia with Lewy bodies presents particular difficulties. Hallucinations and delusions are common in patients with this form of dementia. Sometimes these are not distressing (they may even be pleasant), but sometimes they are very upsetting, with fears of persecution, torture or even murder.

One of the ways in which dementia with Lewy bodies was first recognised as a separate disorder was when it was found that some patients deteriorated or even died when neuroleptic drugs were used. Lewy body inclusions were found in the cortical cells of the brains of these patients at post-mortem.

Ordinarily, neuroleptics might be considered in the treatment of hallucinations or delusions. However, in dementia with Lewy bodies, there is a significant risk of extrapyramidal side-effects, such as Parkinsonism. It is important always to be alert to the possibility of a diagnosis of dementia with Lewy bodies when a person presents with cognitive impairment and psychotic symptoms. Atypical neuroleptics are less likely to cause problems, although patients with dementia with Lewy bodies may tolerate these less well. The adage ‘start low, go slow’ is very important here.


Anyone who consents to something, whether it be taking medicine, going into care or signing a legal document, must be able to understand what it is they are agreeing to, the consequences of doing (or of not doing) it, and be able to remember making their decision. But if someone’s comprehension or memory, is impaired how can this be done?

In practice, the presumption of consent is made unless a patient is deemed as being incompetent. In terms of medication this goes even further - consent is not usually considered unless the patient actively refuses to take it. Can this be right? This means that people with dementia can be given treatments without having the ability to know why they are being given, and what they are for, with few safeguards. In Scotland, the Adults With Incapacity Act 2000 attempts to address this, and similar legislation is being drawn up for the rest of the UK, but it remains to be seen how much of an improvement this will be.

One way of dealing with consent is for people to appoint someone to act for them by granting a power of attorney, to come into effect when they are deemed to be incompetent. This, of course, can be granted only when the ability to do so is present, and requires a degree of forward planning. However, once someone has become incompetent, a court can appoint a guardian, with broadly similar powers to an attorney.

Assessment of dementia

Dementia is often discussed in terms of its being mild, moderate or severe. Relatives, and sometimes patients, themselves ask: ‘How bad is it?’ But how can such a question be answered? In terms of scores on cognitive testing? On the degree of stress to the patients, or to the carers? Or perhaps on the degree of behaviour disturbance? There are no easy answers, and the terms mild, moderate and severe are global ones. Relating them to but one aspect of dementia is meaningless.

Cognitive testing

To confirm a diagnosis of dementia it is essential to demonstrate a change in the patient from a previous level of function: in cognitive ability and in the ability to carry out the activities of daily living. An important part of any assessment, therefore, is the involvement of a relative, friend or carer who knows the patient well.

Rating scales

Various rating scales are used in diagnosing dementia. Perhaps the most widely known is the Mini Mental State Examination (MMSE) (Box 3). This is an old scale but it is still popular, because it makes some attempt to test aspects of cognition in addition to memory. It is also because only minimal training is needed to administer it consistently. A score of less than 26 suggests a possible dementia diagnosis.

The MMSE was designed to screen for the presence of cognitive impairment, but it is now often used to measure change, decline, or improvements - where cognitive-enhancing drugs are used.

Other widely used scales include the Abbreviated Mental Test Score (scoring out of 10), the information and orientation part of the Cambridge Assessment Programme for the Elderly (CAPE), (scoring out of 12) and the ADAS-Cog, a more extensive test developed to assess response to cognitive enhancers.

Although there are scales to measure activities of daily living skills, these are useful only for research purposes; generally, descriptions from a carer are more meaningful. Rating scales are also used in this context, but there is no one scale which is in widespread use. Again their main role is in the field of research.

Physical examination

The advent of drugs for Alzheimer’s makes it important to identify as accurately as possible the actual diagnosis, rather than simply saying the patient has dementia. An essential part of assessment is therefore a physical examination, with supporting tests. Physical examination should look for signs of concomitant illness, and include a neurological examination (the presence of focal signs such as asymmetrical weakness or reflexes might be a pointer to dementia of a vascular origin). In addition to helping make a diagnosis, examination adds to the overall needs assessment.

Biochemical tests

Biochemical and haematological tests are essential. These include a full blood count, erythrocyte sedimentation rate (ESR) and/or c- reactive protein (CRP), B12 and folate estimation, urea and electrolyte levels, liver function tests and thyroid hormone levels. Some authorities feel that patients should also be tested for syphilis, a now rare cause of dementia.

Brain scans

Generally, thorough clinical assessment by an experienced clinician is at least as accurate as a scan when compared with pathological post-mortem (the only absolutely accurate way of establishing the cause). There are, however, some instances when a scan should be considered essential. In general terms, such instances are when the dementing illness has a presentation that is in some way different from what is normally seen. This includes younger age of onset, localised signs or symptoms, associated problems such as incontinence or mobility difficulties, and situations where appropriateness of treatment might be an issue.

Three types of scan are in common use: computerised tomography (CT, or CAT); single photon emission computerised tomography (SPECT), and magnetic resonance imaging (MRI). CT is the most widely used, and is the simplest and cheapest. It provides a structural image of the brain, and is adequate for most purposes. Areas of atrophy can be seen, together with evidence of vascular disease, such as infarcts and thinning of any areas such as the temporal lobe - of particular importance in Alzheimer’s disease. A CT scan shows changes relating to structure rather than function.

SPECT scanning uses similar technology to CT scans, but involves injecting a radioisotope into the bloodstream so that the scanner can assess the blood flow in the brain, which is an indicator of brain activity. SPECT scanning shows images that are typical of Alzheimer’s disease, localising where damage is. It is very useful in diagnosing frontotemporal dementia.

MRI scans use a procedure dependent on the fact that ions (in this case of hydrogen) emit a particular radio frequency when stimulated. Using magnets, the ions are stimulated to move, so that an accurate picture of the brain’s structure is produced. Another investigation that is sometimes used is the electroencephalogram, or EEG. This measures brain electrical activity.

Genetic testing

As yet, genetic testing has little part to play in the assessment process, other than in the case of familial (usually early onset) dementia or in Huntington’s disease. However, with growing awareness of genetic risk, patients or their relatives may wish to be referred to a geneticist for counselling.

Depression and dementia

Pseudodementia is a term with various meanings, but in this context it means the development of apparent cognitive impairment that is, in fact, due to depression. Burns et al. (1990) have noted that depression occurs in 20 per cent of people with dementia. There is evidence that in some patients an episode of depression is a prodromal phase of a dementing illness (Katona, 1995).

How, then, is it possible to distinguish pseudo-dementia from a true dementia? A number of differentiating features is indicated in Box 4.


People with dementia suffer physical and cognitive loss, which may cause them distress, so activating behaviour problems. However, some of these problems may have physical causes, for example, arthritis, and management should consider these possibilities. Confirming a diagnosis of dementia involves cognitive assessment using rating scales; biochemical analysis and, if necessary, brain scans.

Next week: Issues of care.

  • Comment

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Links may be included in your comments but HTML is not permitted.