Over 15% of people on the dementia register were receiving low-dose anti-psychotics. This compares with the NDS, which estimated a point prevalence of 25% [14
] and an observational study, involving a random sample of GP practices across five primary care trusts, which found that 26% of people with dementia were prescribed anti-psychotics [27
]. The recent national audit calculated that in 2011, nationally 6.8% and within Medway PCT 10.5% [Personal Communication with Jonathan Hope] of people with dementia received an anti-psychotic [3
]. However, only 45.7% of GP practices across England participated in the national audit due to technical issues or insufficient resources for data extraction [3
]. Across Medway PCT 17.5% of practices participated in the national audit compared with 98.3% of practices in this project [3
Both the national audit and this project relied upon the accuracy of the dementia registers. However, in many cases a formal diagnosis of dementia may not be recorded [3
]. A typical GP practice would contain between 12 to 20 people with dementia and between three to five people with dementia on low-dose anti-psychotics [14
]. In this project 26 practices failed to report anyone on the dementia register receiving low-dose anti-psychotics. The dementia register was set-up in 2006/07 in Medway and should have been well-established. However, an Alzheimer’s Society national survey identified significant under diagnosis; in Medway only 43.8% of the expected number of people with dementia received a diagnosis, which reflects the national picture [29
]. This confirms findings elsewhere e.g. NHS Dumfries and Galloway estimated that the QOF dementia register only contained 32.9% of the expected dementia population [30
]. Therefore, both the national audit and this project may under-estimate the number of people with dementia prescribed anti-psychotics.
To confirm the extent of the usage of anti-psychotics in dementia, an alternative approach to solely relying on the dementia register may be required. One such alternative option would be to initially identify everyone aged over 65 on low-dose anti-psychotics. A priori criteria for a diagnosis of dementia including a prescription for a cognitive enhancer, in addition to presence on the dementia register, could then be applied. This population identified by the alternative approach should be compared to that identified by solely relying upon the dementia registers, and if it captures the true magnitude of the issue used, as entry criteria for any future intervention.
Prescribing was concentrated in a relatively small number of practices and a person with dementia in a residential home was nearly 3.5 times more likely to receive a low-dose anti-psychotic than someone with dementia living in their own home. This confirms earlier research, which highlighted that anti-psychotic use for people with dementia in nursing homes was becoming increasingly prevalent [31
]. Therefore, to achieve maximum benefit, and in line with the NDS, and the approach adopted in this project, baseline data should be used to target where the intervention is delivered [14
The most commonly used anti-psychotic was amisulpride, followed by risperidone and quetiapine. These three anti-psychotics were also identified by an expert group of old age consultant psychiatrists as the most appropriate treatments for challenging behaviour in dementia [32
]. However, the efficacy of amisulpride for this indication is unproven, and although there may be less harm associated with quetiapine, there is no evidence that quetiapine is effective in people with dementia [22
]. Furthermore, neither amisulpride nor quetiapine are licensed to treat the behavioural and psychological symptoms of dementia. The licensed treatment, risperidone, which is recommended by local and national guidelines, was only used in 23.0% of instances [22
]. Risperidone was one of the first anti-psychotics highlighted by the Committee of Safety on Medicines to cause excess mortality and local and national guidelines, which highlight that all anti-psychotics carry a similar risk, appear not to have been fully communicated to prescribers [24
There are various limitations with the project in addition to issues highlighted regarding the accuracy of the dementia register. The project was solely conducted within a single PCT, and whilst the PCT covered a very mixed area, the results are not necessarily generalisable to other locations. People initiated on anti-psychotics by secondary care were excluded and a joint pharmacist and GP review, linking with secondary care including experts in the non-pharmacological management of BPSD, might have a more significant impact in limiting the inappropriate prescribing of anti-psychotics [31
In the short-term, the pharmacy-led intervention reduced the prescribing of anti-psychotics. However, the scope was relatively limited, and excluded other psychotropics, which might be used in place of anti-psychotics [37
]. Further, patients were not formally followed-up to determine whether the anti-psychotic was stopped, or simply the dose reduced. The symptoms of BPSD fluctuate and continuing input is likely to be required. To demonstrate such a long-term sustainable quality improvement one would require a full-scale research project which was beyond the scope of this evaluation of a clinical service. An initial step would be a pilot randomised controlled trial (RCT) of a specialist pharmacist-led medication review of psychotropics based upon guidelines from the Alzheimer’s Society [22
]. Such a pilot should include a health economic assessment. A positive pilot, and subsequent full trial, could inform future service development.
Key Practice Implications
– Whilst initial data is encouraging further research is required on the role of pharmacy-led interventions to reduce the prescribing of anti-psychotics and other psychotropics for people with dementia.
– Any intervention should focus on care homes where the majority of anti-psychotic prescribing appears to occur.
– The accuracy of the dementia register in identifying people with dementia prescribed low-dose anti-psychotics is unclear.
– Audit data including the recent national audit may under-estimate the usage of anti-psychotics in people with dementia.