Interpretable Machine Learning for Fall Prediction Among Older Adults in China

Abstract

Falls in older adults are potentially devastating while an accurate fall risk prediction model for community-dwelling older Chinese is still lacking. The objective of this study was to build prediction models for falls and fall-related injuries among community-dwelling older adults in China. This study used data (wave 2015 and 2018) on 5,818 participants from the China Health and Retirement Longitudinal Study (CHARLS). 107 input variables at the baseline level were regarded as candidate features. Five machine learning algorithms were used to build the 3-year fall and fall-related injury risk prediction models. SHapley Additive exPlanations (SHAP) was used for the prediction model explanation. Analyses were conducted in 2022. The Logistic regression (LR) model achieved the best performance among fall and fall-related injury prediction models with an area under the receiver operating characteristic curve (AUC-ROC) of 0.739 and 0.757, respectively. Experience of falling was the most important feature in both models. Other important features included basic activity of daily living (BADL), instrumental activity of daily living (IADL), depressivesymptoms, house tidiness, grip strength and sleep duration. The important features unique to the fall model were house temperature, sex and flush toilets, while lung function, smoking and Internet access were exclusively related to the fall-related injury model. This study suggests that the optimal models hold promise for screening out older adults at high risk for falls in facilitated targeted interventions. Fall prevention strategies should specifically focus on fall history, physical functions, psychological factors and home environment.