Risk assessment of wandering behavior in mild dementia.

2016 
OBJECTIVE: This prospective longitudinal study aims to determine the risk factors of wandering-related adverse consequences in community-dwelling persons with mild dementia. These adverse consequences include negative outcomes of wandering (falls, fractures, and injuries) and eloping behavior. METHODS: We recruited 143 dyads of persons with mild dementia and their caregivers from a veteran's hospital and memory clinic in Florida. Wandering-related adverse consequences were measured using the Revised Algase Wandering Scale - Community Version. Variables such as personality (Big Five Inventory), behavioral response to stress, gait, and balance (Tinetti Gait and Balance), wayfinding ability (Wayfinding Effectiveness Scale), and neurocognitive abilities (attention, cognition, memory, language/verbal skills, and executive functioning) were also measured. Bivariate and logistic regression analyses were performed to assess the predictors of these wandering-related adverse consequences. RESULTS: A total of 49% of the study participants had falls, fractures, and injuries due to wandering behavior, and 43.7% demonstrated eloping behaviors. Persistent walking (OR = 2.6) and poor gait (OR = 0.9) were significant predictors of negative outcomes of wandering, while persistent walking (OR = 13.2) and passivity (OR = 2.55) predicted eloping behavior. However, there were no correlations between wandering-related adverse consequences and participants' characteristics (age, gender, race, ethnicity, and education), health status (Charlson comorbidity index), or neurocognitive abilities. CONCLUSION: Our results highlight the importance of identifying at-risk individuals so that effective interventions can be developed to reduce or prevent the adverse consequences of wandering. Copyright © 2015 John Wiley & Sons, Ltd. Language: en
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    34
    References
    29
    Citations
    NaN
    KQI
    []