Establishment of a fall risk prediction model for elderly sarcopenia patients based on SMOTE algorithm
SUN Min
WANG Ya
DING Zuoling
QIAN Weiqun
MENG Ya
Abstract：Objective To explore the risk factors for falls in elderly patients with sarcopenia and construct a risk prediction model based on SMOTE algorithm.Methods A total of 256 elderly patients with sarcopenia admitted to a hospital from December 2020 to Septem⁃ber 2022 were selected as the research subjects,and divided into a fall group and a non-fall group according to the occurrence of falls.Logistic regression analysis was used to screen the risk factors of falls in elderly patients with sarcopenia.The SMOTE algorithm was used to construct a prediction model for falls in elderly patients with sarcopenia,and the prediction efficiency of the prediction model was analyzed.Results A⁃mong 256 elderly patients with sarcopenia,65 patients had falls,and the incidence of falls was 25.39％.Age≥70 years,severe sarcopenia stage,sleep disorders,diabetes,visual impairment and orthostatic hypotension were risk factors for falls in elderly patients with sarcopenia.The original prediction model Logit(P1)=1.057×age+0.808×clinical stage of sarcopenia+0.901×sleep disorder+0.835×diabetes+0.828×visual impairment+1.221×orthostatic hypotension-2.535.The predictive model based on SMOTE algorithm Logit(P2)=1.043×age+0.879×clinical stage of sarcopenia+0.962×sleep disorder+0.717×diabetes+0.810×visual impairment+1.314×orthostatic hypotension-1.445.ROC curve showed that the area under the ROC curve of the P2 model was 0.952(95％CI:0.920,0.972),which was significantly higher than the area under the ROC curve of the P1 model of 0.761(95％CI:0.693,0.828).The calibration curve of the prediction model based on SMOTE algo⁃rithm showed that the predicted value was in good agreement with the actual value.Conclusion Age,clinical stage of sarcopenia,sleep disor⁃ders,diabetes,visual impairment and orthostatic hypotension are risk factors for falls in elderly patients with sarcopenia.The predictive model based on SMOTE algorithm has good predictive efficacy and is helpful for clinical nurses to identify the high-risk group of falls in elderly pa⁃tients with sarcopenia.
Keywords：elderlysarcopeniafallsnursingriskSMOTE algorithmprediction model
Publication Date：2024-10-10
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：5( 899-903 )