Abstract
Background: Doll therapy (DT) is a non-pharmacological intervention for the treatment of the behavioural and psychological symptoms of dementia (BPSD). We designed a single-blind randomized controlled trial of the 30-day efficacy of DT in reducing the BPSD, professional caregivers’ distress and patients’ biomarkers of stress, and in improving the exploration and caregiving behaviours. Methods: We randomly assigned 134 women with moderate-to-severe dementia living in nursing homes (NHs) to a DT intervention (DTI, 67) or a sham intervention with a cube (SI, 67). Results: From the first to the 30th session, the DTI group showed a significant decrease in the Neuropsychiatric Inventory-NH (NPI-NH) total score and in the NPI-NH-Distress score compared to the SI group (both p < 0.001). We observed a greater interest in the doll than in the cube, a greater acceptance of a separation from the nurse among DTI participants, and caregiving and exploratory behaviours towards the doll. There were no differences between the groups in the stress biomarkers. Conclusions: Consistent with attachment theory, our findings support the 30-day efficacy of DT, as this non-pharmacological intervention promotes perceptions of security by creating a situation in which patients feel confident and engaged in a caregiving relationship with the doll and reduces the challenging behaviours that are stressful for professional caregivers.