A randomized controlled trial exploring the effect of music on agitated behaviours and anxiety in older people with dementia.

Objectives: This study, as part of a larger programme of research, sought to investigate the effect that participation in a 40-min live group music programme, involving facilitated engagement with song-singing and listening, three times a week for eight weeks, had on agitation and anxiety in older people with dementia. Methods: A randomized cross-over design, with music and reading control groups, was employed. Forty-seven participants with mild - moderate dementia, from two aged care facilities in Queensland, Australia, were recruited. Participants were assessed three times on the Cohen-Mansfield Agitation Inventory - Short Form (CMAI-SF) and the Rating Anxiety in Dementia Scale (RAID). Results: A sub-analysis of 24 participants attending >/=50% of music sessions found a significant increase in the frequency of verbal aggression over time, regardless of group (F(2,46) = 3.534, p < 0.05). A series of multiple regressions found cognitive impairment, length of time living in the facility and gender to be predictors of agitation overall and by subtype. Conclusion: Participation in the music programme did not significantly affect agitation and anxiety in older people with dementia. Both the music and reading group activities, however, gave some participants a \'voice\' and increased their verbalization behaviour. Agitation was found to be predicted by a number of background factors (namely level of cognitive impairment, length of time in the facility and gender). Future studies would benefit more from in-depth participant assessment prior to study commencement, helping to moderate the influence of low scores, and by undertaking interventions at times when assessed symptoms are most prevalent.