Objectives Literature regarding Montessori-based activities with older adults with dementia is fairly common with early stages of dementia. Conversely, research on said activities with individuals experiencing late-stage dementia is limited because of logistical difficulties in sampling and data collection. Given the need to understand risks and benefits of treatments for individuals with late-stage dementia, specifically regarding their mental and behavioral health, this study sought to evaluate the effects of a Montessori-based activity program implemented in a long-term care facility. Method Utilizing an interrupted time series design, trained staff completed observation-based measures for 43 residents with late-stage dementia at three intervals over six months. Empirical measures assessed mental health (anxiety, psychological well-being, quality of life) and behavioral health (problem behaviors, social engagement, capacity for activities of daily living). Results Group differences were observed via repeated measures ANOVA and paired-samples t-tests. The aggregate, longitudinal results-from baseline to final data interval-for the psychological and behavioral health measures were as follows: problem behaviors diminished though not significantly; social engagement decreased significantly; capacities for activities of daily living decreased significantly; quality of life increased slightly but not significantly; anxiety decreased slightly but not significantly; and psychological well-being significantly decreased. Conclusion Improvements observed for quality of life and problem behaviors may yield promise for Montessori-based activities and related health care practices. The rapid physiological and cognitive deterioration from late-stage dementia should be considered when interpreting these results.

The purpose of this study was to conduct psychometric reevaluation of the Bristol Activities of Daily Living Scale (BADL) among a population logistically difficult to observe beyond cross-sectional analysis: nursing home residents with advanced dementia (AD). Data from observation-based measures were collected by nursing home staff at two intervals within a three-month time frame among 43 residents identified with AD via medical records and nursing home staff. Three broad properties of BADL were examined: factor structure, reliability, and validity. Principal components analysis determined underlying components. BADL internal consistency was assessed by Cronbach's and Guttman coefficients; test-retest reliability was also observed. Convergent validity was assessed by correlating BADL with theoretically linked measures of quality of life (QOL) and social engagement. Compared with the original evaluation, BADL showed inconsistent factor structure at interval 1 but comparable at interval 2. Reliability coefficients at both intervals were robust and comparable to the original evaluation. BADL demonstrated significant convergence with QOL and social engagement. Psychometric potency of BADL was confirmed, suggesting practice applicability with this AD population. Future research calls for further examination of tools to guide effective interventions with this vulnerable population.