Less than half the patients prescribed antipsychotics have a diagnosis of severe mental illness (SMI), a UK primary care study has found.¹

Data were obtained from The Health Improvement Network (THIN), a large database of anonymised patient records from 562 practices, covering 6.2% of the UK population. The study ran from 1 January 2007 to 31 December 2011.

Overall, 47,724 patients were prescribed antipsychotics, of whom 13,941 received first-generation and 27,966 second-generation drugs. Patients aged 80 and above were more likely to receive antipsychotics, incidence rate ratio 2.23 (95% CI: 2.22-2.25) compared with those aged 40-49.

The three most commonly prescribed first-generation antipsychotics were haloperidol, chlorpromazine and trifluoperazine; the proportion of patients with a diagnosis of SMI (psychosis or bipolar disorder) ranged from 27 to 35%. The three most commonly prescribed second-generation antipsychotics were olanzapine, quetiapine and risperidone; the proportion of patients with SMI were 62%, 36% and 46% respectively.

For patients without SMI, the most commonly recorded mental health diagnoses were anxiety, depression, dementia, sleep disorder and personality disorder. For example, the non-SMI patients prescribed risperidone (54%) included 14% with an anxiety code, 22% with depression, 12% with dementia, 11% with sleep disorder and 4% with personality disorder. Around 1 in 10 patients had no recorded mental health diagnosis.

A significant limitation of the study is that non-SMI diagnoses were included irrespective of the time they were entered. Therefore it was not possible to determine the reasons for prescribing antipsychotics, merely the period prevalence of non-SMI diagnoses among those receiving treatment.

Midway through the study period, the DH produced its report The use of antipsychotic medication for people with dementia which recommended a reduction in the rate of use of antipsychotics to a third of its current level.² A subsequent audit found that there had been a 52% reduction in the number of people with dementia prescribed antipsychotic medication from 2008 to 2011.3 This is encouraging, although more needs to be done, and suggests that the findings of the present study with regard to dementia patients are out of date.

Nevertheless, a substantial group of patients (more than a third of the sample) did not have a diagnosis of SMI or dementia. This is surprising as there are few other recognised indications for the use of antipsychotic medication. The British Association for Psychopharmacology depression guideline recommends augmentation with second-generation antipsychotics as one option in unresponsive patients, but only when dose adjustment and switching of antidepressants have been unsuccessful.⁴ Similarly, there are few indications for antipsychotic use in anxiety patients: the strongest evidence of efficacy is restricted to the use of quetiapine as a second-line treatment for generalised anxiety disorder and the augmentation of antidepressant treatment in patients with OCD.⁵ Antipsychotics are not recommended for sleep disorders6 and NICE recommends that they should only be used for short-term crisis management in patients with borderline personality disorder.⁷

I would strongly recommend that we check the records of all our non-SMI patients who are taking antipsychotics. We should review:

• The indication for treatment: is this in accordance with published guidelines?

• The need for continued treatment: do we need to seek advice from secondary care?

• Cardiovascular risk (including BMI, lipids, HbA1c): do the risks outweigh the benefits?