Behavioural symptoms of dementia in residential settings: A selective review of non-pharmacological interventions

Abstract

For people with dementia living in residential settings, behaviours such as aggression, screaming, restlessness, agitation and wandering are a frequent reason for referral to specialist mental health services for older people. Psychosocial models of dementia have grown in prominence and non-pharmacological interventions have been recommended in professional and government policy statements, either as a first line of treatment or alongside medication. Studies of their effectiveness have been criticised for being poorly controlled, focusing on milder behaviour problems and for requiring a disproportionate use of resources. The recent ruling that risperidone and olanzepine should not be used to control behavioural symptoms in dementia makes it timely to review the evidence for alternative treatments. The current review is a selective one of different types of studies including studies of staff training and liaison interventions, studies of a range of different therapeutic interventions and individualized interventions within a single-case methodology. It is argued that different types of research methodology are appropriate for different studies and that there is still too little evidence to provide firm guidelines. In conclusion, a structured decision-making process for selection of interventions is proposed, in which the limited available evidence can be drawn together to provide a basis for targeting clinical resources while the research evidence is strengthened.

Acknowledgements

Thanks are due to Professor Sarah Hampson and Dr Chris Gilleard for comments on an earlier draft of this work and to Dr Victoria Hill for comments on the first draft of this article; also to Professor Esme Moniz-Cook for providing publications in draft cited here.