Longitudinal changes following the introduction of socially assistive robots in nursing homes: a qualitative study with ICF framework and causal loop diagramming

Background: Socially assistive robots introduced in nursing care settings have multidimensional psychological impacts on care recipients and caregivers. This study aims to explore the longitudinal changes induced by socially assistive robots, focusing on a chain of human behaviors. Methods: In this qualitative study, nine participants from two nursing homes who had experience in manipulating socially assistive robots were interviewed in a semi-structured focus group using a topic guide to explore the changes in care recipients and caregivers. Following the framework analysis method, the transcripts were coded using the International Classification of Functioning, Disability, and Health (ICF). The identified codes were charted for each side—care recipients and caregivers sides— using a causal loop diagram, a tool used to visualize nonlinear dynamics in complex systems. Results: Three and seven kinds of codes in the domains of “body functions” and “activities and participation”, respectively, were identified on the care recipients’ side; whereas on the caregivers’ side, one and five kinds of codes in the domains of “body functions” and “activities and participation”, respectively, were found. The codes obtained from the facility with longer experience were represented graphically as a reinforcing feedback loop, in which favorable changes were amplified in a chain of events. Robot use directly changed the mental functions of care recipients, and the caregivers’ perceptions of these positive changes led to their own emotional and behavioral changes, which would reduce the burden of care. Moreover, the findings suggest that sharing information regarding these changes and objectively recognizing the effectiveness of robots among staff members can be the key to continuous robot use in nursing care settings. Conversely, the figure obtained from the novice facility shows fragmented chain relationships of the codes, indicating that all the effects of robot use are recognizable and form a chain reaction after continuous robot use for more than several months. Conclusions: This study revealed important aspects regarding the longitudinal effects of socially assistive robots in nursing care settings. These implications should be broadly implemented for effective robot use and reduction of the burden of care in nursing care settings. Clinical trial number: Not applicable.