Dementia can interfere with the maintenance of social interactions. The ability to participate in social interactions is one of the elements that enables good social health (Hubert et al., Reference Huber2011), and having dementia does not automatically eliminates the person's opportunity to have good social health (Vernooij-Dassen and Jeon, Reference Vernooij-Dassen and Jeon2016). We highlighted in a previous study that people with dementia who did not know each other interacted spontaneously when they were in a stimulating social interaction setting (Mabire et al., Reference Mabire, Gay, Vrignaud, Garitte and Vernooij-Dassen2016). However, a lack of activity and social interaction in nursing homes is still a widespread issue (Harper Ice, Reference Harper Ice2002). Stimulation of social interactions is rarely used as an intervention and social interactions are seldomly used as social health related outcomes.

Recognizing this, we developed a feasibility study to evaluate the immediate effects of one session of an active psychosocial stimulation on social interactions of people with moderate to moderately severe dementia. Thirty-six residents of a French metropolitan nursing home were recruited. The mean age was 88.56 (6.01) and the mean score of the Mini-Mental State Examination (MMSE; Kalafat et al., Reference Kalafat, Hugonot-Diener and Poitrenaud2003) was 13.81 (3.99). These residents received an active psychosocial intervention using the topic of the four seasons as facilitator of group interaction. The intervention was based on the principle of cognitive stimulation therapy and was aligned with the philosophy of Kitwood's person-centered dementia care promoting residents’ positive ability and acknowledging the individual need for social inclusion (Kitwood, Reference Kitwood1997). Participants were asked to talk about the four seasons and to describe all the characteristics they knew about each season. Pictures of landscapes and fruits were used to stimulate and facilitate the conversation. The moderator encouraged residents to interact. To evaluate social interactions, we used an observation grid, the SOBRI (Social Observation Behaviors Residents Index; Mabire et al., Reference Mabire, Gay, Vrignaud, Garitte and Vernooij-Dassen2016). The SOBRI is composed of two components: social interactions between residents and with care staff. The SOBRI presents a good internal consistency with a Cronbach's α of 0.90 for the first component and 0.85 for the second one. Social interactions are scored in terms of presence or absence during eight minutes before and after the meeting. The pre–post comparisons showed a significant improvement (28.72%) of social interactions between residents after the intervention [t = −2.135 (35)/p = 0.040] and a slight non-significant decrease of social interactions with care staff (−2.73%).

These results were promising, but not convincing without the use of a comparison group. We decided to propose a comparison group with passive stimulation in the same controlled social context than the group with active stimulation. Twenty residents from the same nursing home were recruited. The mean age was 89.60 (5.03) and the MMSE mean score was 15.50 (3.12). The comparator group was comparable to the intervention group in terms of socio-demographics and clinical data. These residents were set around a table with newspapers and magazines at disposal. There was no direct intervention of the moderator. The pre–post test comparisons did not show significant differences but social interactions improved between residents (20.39%) and decreased with care staff (−21.420%).

We compared the mean SOBRI scores of the two groups. There were no differences after the session concerning social interactions with residents [F = 0.014 (1)/p = 0.908]. We observed an increase of social interactions in both groups. Concerning social interactions with care staff, a significant difference was observed in favor of the intervention group [F = 5.315 (1)/p = 0.025]. Residents interacted significantly more after the session with the moderator or staff in the intervention group.

This study has several limitations: the study was not a randomized controlled trial, was not blinded, the sample size was too small, and there was on interval of three years between the two groups comparisons. Despite these limitations, the strengths of this study were to test the feasibility to assess the immediate effects of a psychosocial intervention. The positive tendencies provided preliminary insights into the potential immediate effects of active and passive psychosocial stimulation to engage residents with moderate to moderately severe dementia in social interactions. Focus on the immediate effects of psychosocial interventions appears more and more mattering in the person-centered dementia care to allow individual and daily adaptation in care management in all stages of the disease. Moreover, direct observation can be seen as a relevant method in dementia studies to collect spontaneous and ecological observations. Social interactions can be considered as an important modality of measure of social health.

This study can be helpful to professionals in nursing homes and to researchers to engage in a new field of research on social health and on psychosocial interventions in dementia care. The findings contribute to the knowledge on active and passive stimulation of social interaction. Promoting social inclusion in nursing homes is an important challenge in dementia care management since it has immediate implications in the organization of these institutions, requiring adaptations of settings to stimulate social interaction.