Soyinka & Lawley (Psychiatric Bulletin, May 2007, 31, 176-178) report an audit of a crucial treatment in dementia. The finding that 54% of patients received an antipsychotic accords with an audit we conducted in which 49% of patients with behavioural and psychological symptoms of dementia received such treatment. These patients were living in care homes and were referred by general practitioners. We disagree that this represents a ‘ concerning trend of “medicalising”’ such symptoms. These patients are probably at the severe end of a spectrum so it is not surprising that many require antipsychotics. We feel that many patients do benefit from these drugs. Even CATIE-AD (Reference SCHNEIDER, TARIOT and DAGERMANSchneider et al, 2006) report effectiveness of antipsychotics but suggest restriction to patients with minimal side-effects and where benefit is observed.

We disagree with the assumption that findings can be generalised across the country; 50% of patients in the report received typical antipsychotics compared with only 12% in our audit. We note that the National Institute for Health and Clinical Excellence does not distinguish between antipsychotic subtypes (National Institute for Health and Clinical Excellence, 2006) and we wonder if this reflects the growing realisation, referred to by Soyinka & Lawley, that cerebrovascular risk might be shared by all antipsychotics, not just the atypicals.

We have a designated community psychiatric nursing service for care homes, which improves quality of care with targeted antipsychotic use and prompt follow-up and liaison. All patients in our audit had a clear indication for treatment and 94% were reviewed within 3 months. We should use antipsychotic drugs when required to alleviate patient suffering and agree that documented review of patients on antipsychotic drugs is important.