Guidelines for the use and review of antipsychotic drugs need to be urgently reassessed in light of new findings from a study of how the drugs are used in people with dementia in nursing homes.

The research, by scientists from the Institute of Psychiatry, Psychology & Neuroscience (IoPPN) at King's College London, shows that the use of antipsychotic drugs by people with dementia in nursing homes can be effectively reduced through the use of a review protocol, which includes regular scrutiny of prescriptions and targeted education for physicians and nurses.

Importantly, the reduction in use of antipsychotic drugs resulting from use of the review protocol alone appears to lead to a worrying deterioration in patients' neuropsychiatric symptoms (such as aggression or psychosis). However, when the review protocol is accompanied by a programme of social interaction and activities, patients' neuropsychiatric symptoms improve, and mortality rates improve compared to when no review protocol or social interaction intervention is used.

Antipsychotic medication is prescribed to people with dementia to reduce the symptoms of aggression, agitation and psychosis, which often accompany moderate to severe dementia. While the drugs' short-term effect in reducing these symptoms is well established, the benefits of long-term treatment are less clear. Moreover, concerns have been raised about the drugs' long-term safety, with evidence suggesting that prolonged use may result in an increased risk of cognitive decline, stroke, and death.

As a result of these concerns, rates of antipsychotic drug prescribing have declined in recent years, although it is thought that at least one in five people with dementia in nursing homes are still using the medication.

A team of researchers led by Professor Clive Ballard, from the Wolfson Centre for Age-Related Diseases at the IoPPN, King's College London, tested the effects of three interventions - antipsychotic review, social interaction, and an exercise programme. Working with 16 care homes in London, Oxfordshire, and Buckinghamshire, the researchers assigned each of the homes one, or a combination, of the interventions, to be delivered over nine months.

In total, 277 people with dementia took part in the study, and their use of antipsychotic medication, neuropsychiatric symptoms, depression and mortality rates were measured before and after the interventions were delivered.

The results, published in The American Journal of Psychiatry, show that while rates of antipsychotic usage dropped sharply in groups which received the review intervention, this was also accompanied by an increase in neuropsychiatric symptoms - except in the groups which also received the social interaction intervention. In addition to experiencing an improvement in neuropsychiatric symptoms, the mortality rate for patients in this group was substantially reduced compared to those who received no intervention (19% vs. 35%). Exercise programmes also improved neuropsychiatric symptoms, but didn't have any effect on depression.

According to Dr Anne Corbett, a co-investigator on the study, "This study provides clear evidence for the effectiveness of a cheap, practical set of interventions for reducing antipsychotic medication use by dementia patients in nursing homes. Current best practice guidelines for the use of antipsychotic drugs indicate that their long-term use needs to be carefully reviewed, but our results demonstrate the critical importance of ensuring that any reduction or stopping of medication is accompanied by non-drug approaches, particularly social interaction. Reducing the use of these drugs without any other interventions could result in worsening of neuropsychiatric symptoms, which are highly distressing to the person, their carers and loved ones. These findings should prompt an urgent review of guidelines for the use of antipsychotic drugs, to ensure that review of these medicines are in the best interests of people with dementia."

Professor Clive Ballard and Dr Anne Corbett are supported by the NIHR Biomedical Research Centre and Dementia Unit at the South London and Maudsley NHS Foundation Trust and King's College London.