Antipsychotic prescribing for behavioural and psychological symptoms of dementia

AIMS AND METHOD To review the quality of information and advice contained in correspondence 
from old age psychiatrists to general practitioners (GPs) regarding the 
prescription of antipsychotic drugs for the management of behavioural and 
psychological symptoms of dementia. Discharge summaries ( n =22) and 
subsequent out-patient review letters were examined and compared with 
evidence-based guidelines in two phases of an audit cycle; first in 2002 and 
latterly in 2005. RESULTS Practice was below acceptable standards during both phases of the audit 
cycle, with an actual drop in the quality of explicit advice given to GPs in 
2005, despite national publicity about the issues and guidance from the Royal 
College of Psychiatrists. CLINICAL IMPLICATIONS The prescription of antipsychotic drugs is associated with an adverse 
prognosis for people with dementia. As such, it is imperative that such 
treatment is regularly reviewed and time limited. Old age psychiatrists need 
to ensure that this message is communicated to their primary care 
colleagues.

To review the quality of information and advice contained in correspondence from old age psychiatrists to general practitioners (GPs) regarding the prescription of antipsychotic drugs for the management of behavioural and psychological symptoms of dementia. Discharge summaries (n=22) and subsequent out-patient review letters were examined and compared with evidence-based guidelines in two phases of an audit cycle; first in 2002 and latterly in 2005.

RESULTS
Practice was below acceptable standards during both phases of the audit cycle, with an actual drop in the quality of explicit advice given to GPs in 2005, despite national publicity about the issues and guidance from the Royal College of Psychiatrists.

CLINICAL IMPLICATIONS
The prescription of antipsychotic drugs is associated with an adverse prognosis for people with dementia. As such, it is imperative that such treatment is regularly reviewed and time limited. Old age psychiatrists need to ensure that this message is communicated to their primary care colleagues.
It is estimated that over 80% of people with dementia experience one or more recurrent or persistent symptoms such as hallucinations, delusions, aggression, agitation, shouting, screaming, repetitive actions, sleep disturbance or wandering behaviour during the course of their illness (Ballard et al, 1995). These are collectively known as the behavioural and psychological symptoms of dementia. The clinical management may include both psychosocial and pharmacological measures such as the prescription of medication, including antipsychotics. However, use of antipsychotics in older people with dementia can be associated with significant problems, including fractured neck of femur as a result of falls, impaired cognitive functioning and an adverse prognosis (Bouman & Pinner, 2000). Despite these concerns, a number of studies have found excessive use of antipsychotics in nursing home residents (Jenks & Clauser, 1991;Buck, 1998). Experience from the USA has shown that following the introduction of clear guidance supported by legislation, the prescription of such drugs for residents of nursing homes can be reduced by up to 41% (Shorr et al, 1994). Subsequent studies in the UK, utilising the same criteria, suggest that between 54% and 88% of antipsychotic use in residential/nursing homes may be inappropriate (McGrath & Jackson, 1996;Furniss et al, 2000).
Until relatively recently, old age psychiatrists had considered 'atypical' antipsychotics to have a more favourable side-effect profile than the older 'typical' drugs in their patient group. However, in March 2004, the Committee on Safety of Medicines reported an apparent two-to threefold increase in the risk of cerebrovascular events in people with dementia prescribed olanzapine or risperidone, and recommended that these drugs should not be used in this group (message from Professor Gordon Duff, Chairman, Committee on Safety of Medicines; CEM/CMO/2004/1). In response to this advice, a joint guidance note was issued by the Faculty of the Psychiatry of Old Age of the Royal College of Psychiatrists, and other stakeholders, including the Alzheimer's Society and was updated in 2005 (Royal College of Psychiatrists, 2005). This guidance supports the use of antipsychotic drugs for particular behavioural and psychological symptoms, when the problem is severe and when the individual or others may be placed at serious risk as a result of their symptoms. Emphasis is placed on the need to adequately document the reasoning and discussions involved in the decision to prescribe. Further, there is a recommendation that drug treatment is timelimited, with supporting evidence that antipsychotics can be withdrawn successfully in people who have been relatively free from symptoms for 3 months .
Old age psychiatrists have a key role in ensuring that this message is understood by other prescribers, including general practitioners (GPs). Communication of clear guidance between secondary and primary care is therefore essential.
This audit examined the quality of information conveyed from secondary care to primary care in the Hull and East Riding area, when psychiatric in-patients with behavioural and psychological symptoms of dementia, who had been prescribed antipsychotics, were discharged to local residential/nursing homes. A first phase of the audit had taken place in 2002 when poor standards had been identified, and had led to a highlighting of the associated issues at a trust-wide level, including the education of junior doctors (Inasu et al, 2004). This report completes the audit cycle.

Method
A complete list of all people (with functional and organic illness) discharged from the five acute in-patient psychiatric units for older people within Hull and the East Riding of Yorkshire, served by the Humber Mental Health Teaching NHS Trust, was obtained from computerised records, for the period 1 January to 30 June 2005. The trust does not have dedicated in-patient beds for the original papers continued healthcare of people with dementia. People with a diagnosis of dementia who were discharged on at least one antipsychotic medication were identified. The discharge summaries and subsequent out-patient letters for this group were examined for the following information: . Standard 1 Indications for the use of antipsychotic medication. All patients should have at least one of the following recorded in correspondence: (a) an explicit description of the indications for treatment (b) 'target'symptoms and behaviours as described in the Neuropsychiatric Inventory (Cummings et al, 1994) . Standard 2 Review of antipsychotic medications.
(a) all patients should have been reviewed by the psychiatric team within 3 months of discharge, with documented evidence of a review of medication having taken place, or in the absence of this (b) evidence in correspondence of explicit advice being offered to GPs regarding the duration of treatment or the need for the GP to review the prescription of antipsychotic. Review of the patient without documentary evidence of the consideration of medication would not meet this standard.
The initial phase of the audit had taken place in 2002, and therefore pre-dated the recommendations. Hence, slightly different criteria had been used. (The previous standards had stipulated that once the maintenance dose of antipsychotic drug was achieved, there should be a review at least once every 12 months rather than the 3 months now recommended). Nevertheless, very poor standards had been highlighted (see Table 1).

Results
In total 106 people were discharged during the study period, of whom 41 (39%) had a diagnosis of dementia; 22 of the 41 people with dementia (54%) were discharged on at least one antipsychotic medication; no one was discharged on more than one antipsychotic medication; this was an improvement on the 2002 audit where one person had been in receipt of two antipsychotic medications. Table 1 details the findings when agreed standards were applied to discharge summaries and subsequent correspondence to GPs. Comparison of the 2002 and 2005 figures shows that standards had not improved since the first phase of the audit cycle in 2002. Indeed, there had been decline in the quality of advice that had been given to GPs regarding the need to review the prescription of antipsychotic. This seemingly occurred despite the heightened publicity surrounding the issue and despite the presence of clear guidelines.
Of equal interest is the marked alteration in the balance of prescribing typical or atypical antipsychotic drugs, with the prescription of typical drugs increasing from 7% in 2002 to 50% in 2005 (see Table 2). This, presumably, is a consequence of the guidance from the Committee on Safety of Medicines regarding olanzapine and risperidone.
Although not within the scope of the audit, an incidental finding was that a number of patients had been discharged on a benzodiazepine, again without appropriate advice being given on the need for review and monitoring.

Discussion
The focus of this completed audit cycle was to assess the quality of information in correspondence from secondary to primary care regarding the very commonly used, but potentially harmful antipsychotic drugs. Consistent and clear information from secondary care is essential, and may have an influence on the GP's prescribing behaviour.

original papers
Soyinka & Lawley Antipsychotic prescribing in dementia  It is therefore important that the available evidencebased guidelines are fully considered and disseminated. This audit revealed poor standards of practice in this regard, despite the issue of antipsychotic prescribing being highlighted, both locally, by the results of the first phase of the audit cycle, and nationally, by the publicity surrounding the Committee on Safety of Medicines guidance on risperidone and olanzapine in 2004 (message from Professor Gordon Duff, Chairman, Committee on Safety of Medicines; CEM/CMO/2004/1), and the subsequent launch of the Faculty of the Psychiatry of Old Age guidance (Royal College of Psychiatrists, 2005). There was no apparent change in the frequency of prescribing of antipsychotics for older people with behavioural and psychological symptoms of dementia during the audit period, but there was a shift towards prescribing the older 'typical'drugs. Such 'typical' drugs are associated with an increase in disabling sideeffects in older people with dementia, including extrapyramidal and anticholinergic problems, as well as potential worsening of confusion. Moreover, there is a mounting evidence base to suggest that they are likely to increase the risk of cerebrovascular events and death to the same order as their 'atypical' cousins (Gill et al, 2005;Schneider et al, 2005;Wang et al, 2005). It is unlikely that practice within the Humber Mental Health Teaching Trust differs greatly from other old age psychiatry services. Therefore, we assume that the issues can be generalised to many other areas in the country.
Despite the emphasis in this audit on drug treatments, it is clearly important for non-pharmacological management to be considered as first-line treatment. Where available, aromatherapy, reality orientation, validation therapy and recreational activity should be considered as treatment options (Douglas et al, 2004). Despite their widespread use, drug treatments only have a modest supporting evidence base in the management of the behavioural and psychological symptoms of dementia (Sink et al, 2005). Although this audit concerned the management of people who required inpatient care, and are therefore more likely to represent the more severe end of the spectrum, it does confirm the continued and concerning trend of 'medicalising' the behavioural and psychological symptoms by the prescription of drug treatments. More than half of those with dementia were discharged in receipt of an antipsychotic drug, at a time when concern about the effects of such prescribing had received significant publicity.
Following the introduction of the National Service Framework for Mental Health, and the new General Medical Services contract, registers of people with severe and enduring mental health problems are being developed in primary care. Until recently, these have not addressed the needs of people with dementia, a situation reinforced by the initial exclusion of dementia from the Quality and Outcomes Frameworks (QOF) contained in the contract. However, from April 2006, following successful lobbying by the Alzheimer's Society, dementia care became one of nine new areas introduced into the QOF. It is now a target for GP practices to produce a register of all patients diagnosed with dementia, and for a review of their care to take place at least every 15 months. Although the time frame is far from ideal, this does provide a tremendous opportunity to ensure that such a review includes a consideration of the appropriateness of medication, including antipsychotic drugs, based on the evidence-based guidelines (Royal College of Psychiatrists, 2005). Secondary care services have a clear educational role in this regard. It is also important that old age psychiatrists follow good practice models in their written communications with primary care. Indeed, it is time for psychiatrists to grasp the issue and practice what we preach.

Declaration of interest
None.