Purpose: To develop a simple screen based on easily collectable measures to identify older people living in residential care facilities at high risk of falls.
Methods: This prospective study was conducted in seven residential care facilities in the U.K. Residents aged>60 years who were not bedbound or terminally ill participated. Demographics, medical history, medication use, cognition (mini mental state examination (MMSE)), function (Barthel, balance and sit-to-stand ability) and behavior (neuro-psychiatric inventory (NPI) and impulsivity) were recorded at baseline. Falls and injuries were prospectively recorded over 6 months. Data were analyzed for differences between fallers and non-fallers and significant variables entered into logistic regression analysis.
Results: Two hundred and forty residents completed the study. In the follow-up period, 50% fell ≥1 times. Fallers had worse function, cognition, behavior and balance and took more medications. Falling in the past year, walking frame and hypnotic/anxiolytic and anti-depressant medication use were also associated with increased likelihood of falling. Logistic regression identified MMSE<17, impulsivity score ≥ 2, standing balance score<6, requiring a walking frame, falling in the previous year and use of antidepressants and hypnotics/anxiolytics as independent and significant predictors of falls. The area under the receiver operating curve (ROC) for this model was 0.79 (95% CI 0.73-0.84).
Conclusions: This tool comprising multi-factorial measures provides a simple way of quantifying the probability with which a care home resident will fall over a 6-month period. The tool may also assist in guiding the development and targeting of interventions to prevent falls in this group.
Crown Copyright © 2012. Published by Elsevier Ireland Ltd. All rights reserved.