Feasibility analysis of the design of a psychosocial intervention in dementia: Sensitivity analysis

Background A multicomponent staff training intervention was conducted to improve residential dementia care. The purpose of this work is present the simulation effectiveness design over time. Method Pilot study with mixed methods, two-arm randomized controlled clinical outcomes in 7 care homes Mexico City. Staff outcome measures were: Maslach Burnout Inventory (MBI); Approaches Dementia Questionnaire (ADQ); Sense Competence Care (SCIDS). Residents' Quality Life-Alzheimer’s Disease scale (QoL-AD) and Neuropsychiatric Inventory-Nursing Home Version (NPI-NH). Using functions R package powerlmm, we performed computer simulations based on linear changes a period up 24 weeks. measurements were divided into three time points corresponding schedule our (baseline, post-intervention follow-up). In different trials, specified numbers subjects group sizes, including those study, always considering an unbalanced cluster design. We also considered dropout rate participants, which almost 30% focused period, no additional dropouts at follow-up. determine performance longitudinal fewer or unanticipated dropouts. Result 55 residents 126 participants recruited seven homes. found that highest power achieved for MBI: Personal fulfillment SCIDS: Sustaining personhood, only variables significant effect Time × Treatment interaction model inference. Focusing first these assuming specific values each input parameter, see potentially capable detecting size growth effects intervention. Thus, estimated sample 80% would be nearly 85 corresponds complete data (i.e., dropouts), about 138 90%. Conclusion higher level includes more sufficient achieve effective estimates above threshold while maintaining lowest number participants.