People with dementia living in residential care often face the dual disadvantage of increasing difficulty with communication and reduced opportunities for conversation. Social interaction is central to wellbeing of residents with dementia, so it is important that care staff have the skills to engage in conversation with them. We studied conversations in 20 care staff–resident dyads, to examine conversation structure and content, patterns of engagement within conversations, including the topics around which engagement occurred, and communication behaviours by care staff that appeared to facilitate (or impede) participation by residents. The transcripts were analysed using Discursis, a computational information visualisation tool that allows interactive visual inspection, in context, of the contributions by each speaker, the turn-taking dynamics, and the content recurring within and between speakers. We present case examples (a) where care staff did most of the talking, initiated topics and were responsible for most recurrence of content; (b) where talk was more evenly shared between partners, with some topics initiated and/or elaborated by participants with dementia; and (c) where participants with dementia talked most, with care staff supporting the conversation. We identified accommodative strategies used by care staff, such as reflecting back the other person's responses to sustain engagement. We also noted care staff behaviours that impeded communication, such as not listening attentively and not allowing sufficient time for responses. The results from this study highlight aspects of social communication within the aged care context and suggest ways in which rewarding interactions between staff and residents with dementia might be encouraged.

Background: People with dementia have a range of needs that are met by informal caregivers. A DVD-based training program was developed using research-based strategies for memory and communication in dementia. The effectiveness of the training on the caregiver experience and the well-being of the person with dementia was evaluated.

Methods: A pre-test/post-test controlled trial was undertaken with caregiver–care-recipient dyads living in the community. Measures of the carers’ knowledge of memory and communication strategies, burden, positive perceptions of caregiving, and perceptions of problem behaviors were taken pre- and three months post-intervention. The depression and well-being of the person with dementia were also evaluated. Satisfaction with the training and feedback were measured.

Results: Twenty-nine dyads (13 training group, 16 control group) participated. Bonferroni's correction was made to adjust for multiple comparisons, setting α at 0.00385. A significant improvement was found in caregivers’ knowledge for the training group compared to the control group (p = 0.0011). The training group caregivers reported a reduction in the frequency of care recipient disruptive behaviors (p = 0.028) and increased perceptions of positive aspects of caregiving (p = 0.039), both at a level approaching significance. The training group care recipients had increased frequency of verbally communicated depressive behaviors at a level approaching significance (p = 0.0126). The frequency of observed depressive behaviors was not significantly different between groups.

Conclusions: This approach to training for caregivers of people with dementia appears promising for its impact on knowledge and the caregiving experience. Further research could monitor the impact of the training on broader measures of depression and well-being, with a larger sample.