Exploring the effect of implementing affordable socially assistive pet robots in eight care homes before and during the COVID-19 pandemic: a stratified cluster randomised controlled trial and mixed-method study.

Abstract

Abstract Background Robot pets may assist towards challenges of supporting an aging population with growing dementia prevalence. Prior work focused on impacts of robot seal Paro on older adult wellbeing, but recent studies suggest good acceptability and implementation feasibility of more affordable devices (Joy for All (JfA) cats and dogs), yet effectiveness research was limited. Methods We conducted an eight-month, stratified, cluster randomised controlled trial, in eight care homes in Cornwall, UK. Over four months, four care homes each received two JfA devices (one cat and dog), and four homes received care as usual (intervention and control group). Psychometrics were collected pre and post intervention, to compare change from baseline to follow-up in the intervention vs control group. In the final four months, all eight care homes had devices, but only qualitative data was collected, due to Covid-19 and reduced capacity. The primary outcome was neuropsychiatric symptoms (Neuropsychiatric Inventory – Nursing Home version (NPI)). Care provider burden was a secondary outcome (occupational disruptiveness NPI subscale), alongside the Challenging Behaviour scale, Holden Communication scale, Campaign to End Loneliness questionnaire and medication use. Qualitative data was collected through care staff observation calendars and end-of-study interviews to understand use, experience and impact. We also collected demographic data and assessed dementia severity. In total, 253 residents had robot interaction opportunities, and 83 were consented for direct data collection. This trial was pre-registered on Clinicaltrials.gov (https://clinicaltrials.gov/ct2/show/NCT04168463), and is reported following the CONSORT 2010 statement: extension to cluster randomised trials. Results There was a significant difference in total change from baseline to follow-up for NPI (p=.000) and occupational disruptiveness (p=.031) scores between the intervention and control group. Neuropsychiatric symptoms increased in the control group, while decreasing in the intervention group. No significant difference was seen for communication issues or challenging behaviour. On NPI sub-domains, there was a significant difference from baseline to follow-up for delusions (p=.034), depression (p=.010), anxiety (p=.001), elation (p=.023) and apathy (p=.009), all of which decreased in the intervention group and increased slightly in the control group. The summative impact results suggested most residents who interacted with robots received a positive impact (85%, 46/54). Those who interacted had significantly higher dementia severity scores (p=.001) than those who did not interact. The qualitative results suggested good adoption and acceptability, suitability for subjectively lonely individuals, lack of novelty effect through sustained use and demonstrated ‘reasons for use’ of robots were entertainment, anxiety and agitation. Conclusion Affordable robot pets hold potential in improving wellbeing for care home residents and people with dementia, including reduced neuropsychiatric symptoms and occupational disruptiveness. This work suggests no novelty effect, and contributes towards understanding robot pet suitability, finding interactions were more common among residents with more moderate/severe dementia and potentially those subjectively lonely.