Guidelines on antipsychotics for dementia: are we losing our minds?

Aims and Method The National Institute for Health and Clinical Excellence (NICE) and the Royal College of Psychiatrists have each issued guidance on the use of antipsychotics for behavioural and psychiatric symptoms of dementia (BPSD). We sent all old age psychiatrists an anonymous questionnaire asking for their opinions on these documents and for details of their use of antipsychotics for BPSD. Results The response rate was 202 out of 648 (31.2%). the two documents, though similar in content provoked very different responses, with the College guidance being much more favourably received. All respondents prescribed antipsychotics for BPSD, most commonly quetiapine. Clinical Implications When prescribing antipsychotics for behavioural and psychiatric symptoms of dementia, psychiatrists should take both NICE and College guidelines into account and use their clinical judgement.

team, as such organisations are unlikely to have sufficient familiarity with the case in order adequately to address the needs of the child or be sufficiently knowledgeable about issues of confidentiality. Patients and families may resist support from cultural organisations for fear of the local community becoming aware of their situation. Stigma and discrimination surrounding mental illness are prevalent and this may be more so among ethnic minority groups (Office of the Deputy Prime Minister, 2004).
The children of forensic in-patients are a difficult population group to access. Although we are encouraged to consider 'patients as parents', guidance given to forensic psychiatrists places an emphasis on risk assessment and working with the parent, but not the child (Royal College of Psychiatrists, 2002). Even so, although child protection procedures and the best interests of the child are central to the regulation of visits to patients in secure psychiatric units, it would seem that the responsibility of forensic mental health services towards the children may extend beyond this, and that offering support where possible is good practice. A collaborative approach involving other agencies may be the best way to undertake this.

Declaration of interest
None.
References FALLON, P., BLUGLASS, R., EDWARDS, B., et al (1999) Report of the Committee of Inquiry into the Personality Disorder Unit, Ashworth Special Hospital (vol.
Guidelines on antipsychotics for dementia: are we losing our minds?

AIMS AND METHOD
The National Institute for Health and Clinical Excellence (NICE) and the Royal College of Psychiatrists have each issued guidance on the use of antipsychotics for behavioural and psychiatric symptoms of dementia (BPSD). We sent all old age psychiatrists an anonymous questionnaire asking for their opinions on these documents and for details of their use of antipsychotics for BPSD.

RESULTS
The response rate was 202 out of 648 (31.2%). The two documents, though similar in content provoked very different responses, with the College guidance being much more favourably received. All respondents prescribed antipsychotics for BPSD, most commonly quetiapine.

CLINICAL IMPLICATIONS
When prescribing antipsychotics for behavioural and psychiatric symptoms of dementia, psychiatrists should take both NICE and College guidelines into account and use their clinical judgement.
Individuals with dementia commonly manifest behavioural and psychiatric symptoms of dementia (BPSD) such as aggression, agitation, psychotic symptoms, sleep disturbance and wandering. The off-label use of antipsychotic drugs to treat BPSD is widespread. Studies from around the world have reported that 25-40% of elderly residents in long-term institutional care receive antipsychotics (Hagen et al, 2005;Snowdon et al, 2005;Alanen et al, 2006;Rochan, 2007). A UK survey found 32% of patients in care homes with dementia were prescribed antipsychotics (Alldred et al, 2007). In 2004, the Committee on Safety of Medicines warned of the increased risk of cerebrovascular events with risperidone and olanzapine and said these drugs should not be used to treat behavioural problems in older adults with dementia (Committee on Safety of Medicines, 2004). The Royal College of Psychiatrists' Faculty of the Psychiatry of Old Age and other stakeholders responded by issuing guidance laying down good practice guidelines for the use of antipsychotics in the treatment of BPSD (Working Group, 2004). Further guidance from the Royal College of Psychiatrists was later thought necessary due to reports of inappropriate interpretations of the earlier documents, for example patients having their antipsychotics abruptly withdrawn (Royal College of Psychiatrists, 2005). More recently, the National Institute for Health and Clinical Excellence (NICE) has issued a national clinical practice guideline on dementia with firm recommendations about the pharmacological treatment of dementia (National Collaborating Centre for Mental Health, 2007).

Chao & Kuti Supporting children of forensic in-patients original papers
What do old age psychiatrists feel about all this guidance, some of it highly specific and prescriptive? Is it applicable to the real world of everyday National Health Service (NHS) practice? Does it leave enough room for clinicians to exercise their clinical judgement or do psychiatrists feel they have been left without support when they prescribe antipsychotics? We surveyed the views of career old age psychiatrists to see whether or not they considered the most recent Royal College of Psychiatrists' document (forthwith referred to as the RCPsych 2005 guidance) supportive of their practice and the NICE dementia guideline too restrictive. We wanted to know whether any psychiatrists thought antipsychotics should never be used for BPSD. We sought to determine the percentage of patients with BPSD for whom old age psychiatrists prescribe antipsychotics and which drugs they prescribe most frequently.

Method
We talked to a number of old age psychiatrists to determine their views about the two sources of guidance. Although a range of opinions were expressed, the overall feeling was that the RCPsych 2005 guidance struck a better balance over the use of antipsychotics for BPSD than the NICE dementia guideline. Very few psychiatrists were of the opinion that antipsychotics should never be used. We set out to confirm or refute these views in the questionnaire, which was first piloted in our own hospital. In May 2007, after obtaining approval from the College, we sent a single mailing of an anonymous postal questionnaire to the 608 consultants and 40 associate specialists listed by the College as specialising in old age psychiatry. The questionnaire asked respondents to rate on a ten-point scale (1 -strongly agree, 10 -strongly disagree) their agreement/disagreement with the following statements. Respondents were invited to comment on the NICE and the RCPsych 2005 guidance on dementia and to indicate the percentage of their patients with BPSD for whom they prescribe antipsychotics. They were also asked about the clinical circumstances under which they prescribe antipsychotics for BPSD and the three antipsychotics they use most commonly (with dosage ranges). (The questionnaire is available from the corresponding author on request.) Numerical analyses were conducted using SPSS version 14.0 for Windows. Participants' comments were subjected to qualitative analysis for commonly occurring themes. Two of the authors independently derived a list of topics from the comments and then met to reach a consensus on the major themes.

Results
Of the 648 questionnaires sent out, 207 were returned, of which 202 (31.2%) had been completed. Almost all respondents (176 of 202, 87.1%) worked in NHS community and in-patient settings, 14 (6.9%) worked in NHS community services and most of the rest (n=8, 4.0%) in specialist services.
The majority (130 of 197, 66.0%) of respondents thought the NICE guideline on dementia placed clinically inappropriate restrictions on the use of antipsychotics in the treatment of BPSD. Median score on the scale of 1-10 (1 -strongly agree, 10 -strongly disagree) was 4 (interquartile range IQR=3-7).
A total of 162 comments on the NICE guideline on dementia were received: 40 positive (24.7%), 111 negative (68.5%) and 11 in which the respondent had not read or could not recall the guidance (6.8%). Comments fell into several broadly defined themes (Table 1). Of the 135 comments on the RCPsych 2005 guidance, 105 were positive (77.8%), 20 were negative (14.8%) and 10 had not read or could not remember the document (7.4%). Overall, the NICE guideline on dementia was perceived as too restrictive for secondary care, as placing too great an emphasis on the risks of antipsychotics in BPSD and unrealistic given current resources. The RCPsych 2005 guidance was seen as more balanced, practical and useful to justify prescribing decisions.
Almost all respondents (190 of 199, 95.5%) disagreed with the statement that antipsychotics should never be prescribed for patients with any type of dementia. Median score on the 1-10 scale (1 -strongly agree, 10 -strongly disagree) was 10 (IQR=9-10). All respondents reported that they prescribed antipsychotics for BPSD. The median percentage of their patients with BPSD for whom they prescribed antipsychotics was 40% (range 5-90). The most common indications for antipsychotics were: psychosis (93.3%), aggression (89.1%), agitation (72.0%) and sexual disinhibition (50.3%). Other indications included: when other measures (including non-pharmacological ones) had failed, risk of harm to self or others, severe intractable distress and where a patient's placement was at risk. Some emphasised the importance of first trying other measures before antipsychotics, for example antidepressants and nonpharmacological interventions, and the need to balance risks against benefits. Quality of life was also important: 'the angry, distressed, deluded, unhappy patient would not willingly choose to be like that', 'in dementia it is quality of life that is important, not quantity'.
The three most commonly prescribed antipsychotics for BPSD are given in Table 2, together with the median usual dosage ranges. Quetiapine was by far the most commonly prescribed antipsychotic. Although atypicals were used more often than typicals, haloperidol was the second most popular antipsychotic and several other typicals were also prescribed, albeit less frequently.

Discussion
In this survey of old age psychiatrists, two-thirds of respondents thought the NICE guideline on dementia was too restrictive, whereas over three-quarters felt the RCPsych 2005 guidance supported psychiatrists in prescribing these drugs to individuals with dementia. The NICE guideline provoked a larger number of comments (mostly negative) compared with the RCPsych 2005 guidance which received mostly positive comments. All respondents reported that they prescribed antipsychotics for at least some of their patients with BPSD (on average 40%) and the overwhelming majority disagreed with the original papers Haw et al Antipsychotics for dementia  statement that antipsychotics should never be prescribed for BPSD. Quetiapine and haloperidol were the most frequently prescribed antipsychotics.

Limitations
The study has a number of limitations. The response rate to the questionnaire was 31%, which was low (although not unusual for a single mailing of an anonymous questionnaire) and could reflect sample bias. It may be that those psychiatrists who responded were more negative in their attitudes towards the NICE guideline on dementia and/or more frequently prescribed antipsychotics for BPSD. The questions we asked were designed to be provocative but are likely to have introduced bias. Another possible source of bias was the categorisation of comments as either positive or negative and by theme, even though ratings were made independently by two authors. The fact that quetiapine was the most commonly prescribed antipsychotic is likely to relate to the Committee on Safety of Medicines (2004) alert about risperidone and olanzapine. There is now evidence that the increased risk of death applies to all atypicals (Schneider et al, 2005). Several respondents thought the NICE guideline on dementia encourages the prescribing of the older typical drugs. In our survey, haloperidol was very commonly prescribed and some respondents reported using other typicals. Several studies have reported the risk of cerebrovascular events and increased mortality to be similar for typicals and atypicals (e.g. Gill et al, 2005;Trifiro' et al, 2007), with two recent studies reporting a higher risk of death in patients receiving typicals than atypicals (Wang et al, 2005;Gill et al, 2007). Some respondents pointed out that the increased risk of cerebrovascular events and death with antipsychotics is likely to be a class effect. Evidence of efficacy in BPSD, though limited, is greatest for risperidone and olanzapine, the antipsychotics featured in the Committee of Safety of Medicines alert (Lee et al, 2004).
Both the NICE guideline and the RCPsych guidance discuss the use of antipsychotics in BPSD and appear to come to similar conclusions about the use of these drugs in dementia, yet the psychiatrists in our survey perceived the NICE guideline in a much more negative light. This may be due to the following features of the NICE document: its tone ('people with Alzheimer's disease, vascular dementia or mixed dementias with mild-to-moderate non-cognitive symptoms should not be prescribed antipsychotic drugs'), source (large organisation set up by the Government) and content (inclusion of nonpharmacological measures lacking an evidence base, for example animal-assisted therapy and massage). However, NICE has pointed out their guidance is not binding on clinicians, although it should be taken into account when making clinical decisions (Anonymous, 2007). Many old age psychiatrists appear to be troubled by the NICE guideline on dementia, although they feel the RCPsych 2005 guidance goes some way to redressing the imbalance. Of course, there are other treatments for BPSD which should be considered, such as cholinesterase inhibitors for Lewy body dementia, as well as nonpharmacological interventions.

Declaration of interest
None.