Construction and validation of a risk prediction model for oral frailty in elderly cancer patients
LU Cheng-qian
LU Qin-yi
JIN Xue-qin
SONG Juan
Abstract：Objective To investigate the independent predictors of oral frailty among hospitalized elderly cancer patients develop a predictive model and draw a nomogram,and simultaneously validate the predictive performance of the model.Methods Employing a method of convenience sampling,406 senior cancer patients from hospitals in two advanced facilities in Suzhou City were selected during May 2024 to August 2024.The patients were divided into modeling group(295 cases)and validation group(111 cases).Logistic regression was utilized to investigate the factors leading to oral frailty in older cancer patients.The risk prediction model and nomograms model were constructed using R software,and the area under the receiver operating characteristic(ROC)curve was used to test the prediction effect of the model.Results Univariate analysis revealed that age,marital status,educational attainment,radiotherapy,chemotherapy,grip strength,the scores of geriatric self-efficacy scale for oral health,oral mucositis,poor oral health,and nutritional risk significantly influenced the development of oral frailty in elderly cancer patients(P＜0.05).Logistic regression analysis demonstrated that age,radiotherapy,oral mucositis,reduced grip strength,poor oral health,low oral health self-efficacy,and high nutritional risk as independent predictors of oral frailty in this population(P＜0.05).A nomogram was constructed to predict oral frailty risk in elderly cancer patients.The Hosmer-Lemeshow test demonstrated excellent model calibration(x2=0.439,P=0.932).The model showed outstanding discriminative ability with an area under the curve(AUC)of 0.954(95％ CI:0.934-0.974)in the training cohort and 0.977(95％ CI:0.957-0.998)in the validation cohort.Conclusion The nomogram model constructed based on advanced age,radiotherapy,grip strength,oral mucositis,nutritional risk,oral health-related self-efficacy,and poor oral health exhibits excellent discrimination and calibration.It can assist medical staff in screening high-risk groups of elderly cancer patients for oral frailty,adopt appropriate intervention measures,and decrease the occurrence risk of oral frailty.
Keywords：cancerelderlyoral frailtypredictive modeling
Publication Date：2025-05-20
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：9( 210-218 )