Purpose of the study: A novel logistic regression tree-based method was applied to identify fall risk factors and possible interaction effects of those risk factors.
Design and methods: A nationally representative sample of American older adults aged 65 years and older (N = 9,592) in the Health and Retirement Study 2004 and 2006 modules was used. Logistic Tree with Unbiased Selection, a computer algorithm for tree-based modeling, recursively split the entire group in the data set into mutually exclusive subgroups and fit a logistic regression model in each subgroup to generate an easily interpreted tree diagram.
Results: A subgroup of older adults with a fall history and either no activities of daily living (ADL) limitation and at least one instrumental activity of daily living or at least one ADL limitation was classified as at high risk of falling. Additionally, within each identified subgroup, the best predictor of falls varied over subgroups and was also evaluated.
Implications: Application of tree-based methods may provide useful information for intervention program design and resource allocation planning targeting subpopulations of older adults at risk of falls.