Potential of Snoezelen room multisensory stimulation to improve balance in individuals with dementia: a feasibility randomized controlled trial

Objective: To investigate the influence of multisensory stimulations in a Snoezelen room on the balance of individuals with dementia. Design: Randomized controlled trial. Setting: Canadian long-term care home. Participants: Twenty-four residents (average age 86 years), in a long-term care home diagnosed with dementia, were assigned randomly to intervention and control groups. Nineteen participants completed the study. Interventions: Nine intervention group participants completed 30-minute Snoezelen room sessions twice a week for six weeks. Sessions were guided by participants' preferences for stimulation. Interactions with tactile, visual and proprioceptive sensations were encouraged. Ten control group participants received an equal amount of volunteer visits. Main outcome measures: The Functional Reach Test, the eyes-open Sharpened Romberg and the Timed Up and Go Test with and without dual task, assessed static and dynamic balance at baseline and after the intervention. Falls frequencies were recorded six weeks before, during and after intervention. A journal was kept of observations in Snoezelen room. Results: Split-plot MANOVA analyses revealed no significant effects of unstructured Snoezelen room sessions on participants' balance. There were no multivariate effects of time (F(4,14) = 1.13, P = 0.38) or group (F(4,14) = 0.63, P = 0.65). Group membership did not alter falls frequency. However, observations of participants' interactions with elements of the Snoezelen room, such as imagery-induced head and eye movements, vibrating sensations and kicking activities, captured events that can be used to create specific multisensory balance-enhancing stimulations. Conclusions: Although the null hypothesis was not rejected, further investigation of a potential to influence balance in individuals with dementia through Snoezelen room intervention in long-term care homes is warranted.