Objective To construct a predictive model for the risk of frailty in elderly patients with knee osteoarthritis in the community.Methods A total of 128 elderly patients in the community from March 2021 to May 2023 were selected as the subjects,all of whom were evaluated for knee osteoarthritis weakness using the frailty phenotype(FP)scale and divided into a frail group and a non frail group.Review two sets of data and complete single factor and multivariate logistic regression analysis on the possible influencing factors of frailty in elderly patients with knee osteoarthritis in the community.Based on the results of multiple factors,a column chart model for the occurrence of weakness in elderly knee osteoarthritis patients in the community was constructed using the RMS program in R3.5.3 software,and its predictive value was analyzed.Results Among the 128 elderly patients with knee osteoarthritis in the community,there were 47 cases of weakness,with an incidence rate of 36.72%.The results of univariate and multivariate logistic regression showed that disease course,visual and auditory impairment,exercise,number of comorbid chronic diseases,and social support were independent risk factors for the occurrence of weakness in elderly patients with knee osteoarthritis in the region(P<0.05).Construct a frailty column chart model for elderly patients with knee osteoarthritis in the community.The ROC curve results show that the AUC value of the column chart model for predicting knee osteoarthritis frailty in elderly patients in the community is 0.876,with a 95%CI of 0.813-0.951,an optimal cutoff value of 87.51 points,a sensitivity of 83.17%,and a specificity of 78.48%.Conclusion The incidence of frailty in elderly patients with knee osteoarthritis in the community is relatively high,and there are many influencing factors.Constructing a column chart model for predicting frailty in elderly patients with knee osteoarthritis in the community has high predictive efficacy.
Community elderlyKnee osteoarthritisWeaknessMultivariate logistic regression analysisRisk prediction model