Informative and semantic language features of people with dementia displayed during reminiscence therapy

Communication is essential for people with dementia (PWD) to independently perform activities of daily living. While cognitive training could improve communication abilities in PWD, there is limited evidence of the effect of reminiscence therapy (RT) on communication. Therefore, this study examines this effect, focusing on language features.This study used data from randomised controlled trials applying RT to women with dementia registered in a day-care center. RT using digital material or storytelling was conducted with PWD for 4 weeks. Outcome measures included general characteristics and communicative ability, such as the number of correct information units (CIUs, a measure of informativeness in discourse), the Holden Communication Scale (HCS) score, and verbal engagement. Repeated-measures multivariate analysis of variance and the paired t-test were performed to confirm the effect of RT on communicative ability.There was a significant change over time in the ratio of CIUs (F = 4.35, P = 0.016). In the first 2 weeks of the intervention, there were significant differences in the ratio of CIUs (t = -3.00, P = 0.005), total HCS score (t = 2.28, P = 0.028), and conversation score (HCS subdomain) (t = 2.44, P = 0.019) between two measurement time points (T0, T1), while the number of syllables did not significantly change (t = -1.60, P = 0.117).Informative/semantic linguistic ability was significantly increased in PWD, while the amount of discourse did not change after RT. This result may be helpful to improve effective communication with PWD and train healthcare staff working with PWD.