Data mining and analysis of fall events in hospitalized cancer patients
TAN Rongyan
HU Meihua
PENG Siyi
LI Xuying
HU Yonghong
Abstract：Objective:To explore the characteristics and correlations of fall events in hospitalized cancer patients,providing a theoretical basis for developing fall prevention strategies.Methods:Data mining principles and the Apriori algorithm were used to mine data from 138 adverse fall events reported by a tertiary grade A hospital in Hunan province from April 2022 to July 2024.The effectiveness of association rules was evaluated based on screening criteria and the chi-square test.Finally,strong association rules were determined based on clinical significance.Results:A total of 11 strong association rules were obtained,including stage Ⅳ cancer,distant metastasis,no use of assistive devices before the fall,combined treatment with multiple drugs,malnutrition,cancer-related fatigue,two nurses on duty,etc.Conclusion:Falls among hospitalized cancer patients exhibit certain characteristics.A fall risk early warning model for hospitalized cancer patients should be constructed to dynamically assess and effectively manage their fall risk.Special attention should be paid to patients undergoing combination therapy with multiple drugs,providing nutritional support to further reduce the occurrence of falls.
Keywords：hospitalized cancer patientsfallsassociation rulesdata miningrisk factor analysis
Publication Date：2025-12-15
Online Publishing Date：2025-12-22(First online date of this platform, not the publication date of the document)
Pages：5( 4411-4415 )