To examine effects on symptoms of agitation and depression in nursing home residents with moderate to severe dementia participating in a robot-assisted group activity with the robot seal Paro.
A cluster-randomized controlled trial. Ten nursing home units were randomized to either robot-assisted intervention or a control group with treatment as usual during 3 intervention periods from 2013 to 2014.
Ten adapted units in nursing homes in 3 counties in eastern Norway.
Sixty residents (67% women, age range 62-95 years) in adapted nursing home units with a dementia diagnosis or cognitive impairment (Mini-Mental State Examination score lower than 25/30).
Group sessions with Paro took place in a separate room at nursing homes for 30 minutes twice a week over the course of 12 weeks. Local nurses were trained to conduct the intervention.
Participants were scored on baseline measures (T0) assessing cognitive status, regular medication, agitation (BARS), and depression (CSDD). The data collection was repeated at end of intervention (T1) and at follow-up (3 months after end of intervention) (T2). Mixed models were used to test treatment and time effects.
Statistically significant differences in changes were found on agitation and depression between groups from T0 to T2. Although the symptoms of the intervention group declined, the control group's symptoms developed in the opposite direction. Agitation showed an effect estimate of -5.51, CI 0.06-10.97, P = .048, and depression -3.88, CI 0.43-7.33, P = .028. There were no significant differences in changes on either agitation or depression between groups from T0 to T1.
This study found a long-term effect on depression and agitation by using Paro in activity groups for elderly with dementia in nursing homes. Paro might be a suitable nonpharmacological treatment for neuropsychiatric symptoms and should be considered as a useful tool in clinical practice.