Abstract
Date Presented 04/21/2023
Some physical and mental health variables were identified as important factors of falls among older adults. This study investigated the risk factors that are associated with falls among older adults living alone using machine learning algorithms.
Primary Author and Speaker: Suyeong Bae
Contributing Authors: Ickpyo Hong, Daewoo Pak
PURPOSE: Adults who live alone usually care for themselves independently, and thus, they are more likely to fall and susceptible to injuries. This study aims to evaluate risk factors that are associated with falls and developed a prediction model of falls among older adults living alone using machine learning approaches.
DESIGN: Cross-sectional study.
METHOD: We extracted the data of 803 older adults living alone from the 2020-2021 National Health and Aging Trends Study (NHATS). The variable about whether or not they experienced a fall during 2020 was considered as an outcome. The possible risk factors for falls were selected among demographic, functional, and environmental variables that were available in the 2020 NHATS. We developed the prediction models using the random forest (RF) and extreme gradient boost (XGBoost) and evaluated the models with 5-fold cross-validation, where model performances were compared with model’s accuracy, precision, recall, and area under the curve (AUC).
RESULT: Within 213 older adults who live alone and experienced falls in 2020, 151 were females (70.89%). Among the prediction models, the model from XGBoost showed better performance in terms of accuracy, precision, and recall, while the RF yielded higher values in AUC. The factors that are highly associated with falls were age, cognitive function, and problem in balance for XGBoost; and age, cognitive function, and depression for the RF.
CONCLUSION: From our prediction models, age and cognitive function were the common risk factors that affect the risk of falling among older adults living alone. Thus, it is necessary to develop fall prevention program and policy that efficiently reduce the risk of falls in older adults whose cognitive and functional decline.
IMPACT STATEMENT: Our results suggest that occupational therapists or practitioners would need to consider cognitive function and mental health when they establish interventions to prevent falls in older adults living alone.
References
Elliott, S., Painter, J., & Hudson, S. (2009). Living alone and fall risk factors in community-dwelling middle age and older adults. Journal of Community Health, 34(4), 301–310. https://doi.org/10.1007/s10900-009-9152-x
Lage, I., Braga, F., Almendra, M., Meneses, F., Teixeira, L., & Araujo, O. (2022, in press). Falls in older persons living alone: the role of individual, social and environmental factors. Enfermería Clínica (English Edition). https://doi.org/10.1016/j.enfcle.2022.04.003
