The influencing factors of sarcopenia in elderly people and the construction of a column chart prediction model

Liang Bijuan; Su Jianpei; Gu Jun; Wang Zhimin

doi:10.3969/j.issn.1671-6450.2025.04.017

Abstract：Objective To explore the influencing factors of sarcopenia in elderly people and construct a column chart prediction model based on the influencing factors.Methods Clinical data of 200 hospitalized elderly patients admitted to the Department of Geriatrics of our hospital from January 2021 to July 2024 were collected.According to the occurrence of sarcopenia,they were separated into a sarcopenia group of 60 cases and a non-sarcopenia group of 140 cases.Multivari-ate logistic regression analysis was applied to screen the influencing factors of sarcopenia in elderly people.The rms pack-age in R software was applied to construct a column chart prediction model for the occurrence of sarcopenia in elderly peo-ple.Internal validation of the column chart prediction model was performed using Bootstrap method with 1000 repeated sam-plings.Calibration curve,ROC curve,and clinical decision curve were applied to verify the consistency,discriminability,and clinical value of the column chart prediction model.Results Compared with the non myopenia group,the myopenia group has a high proportion of people aged ≥ 75 years,a low BMI,a high proportion of osteoporosis,a high proportion of diabe-tes,low albumin,and high TG/HDL-C,with statistically significant differences(x2/t=15.629,7.095,14.618,15.126,7.968,7.658,P＜0.001).Age ≥75 years old,osteoporosis,diabetes,high TG/HDL-C are independent risk factors for sarcopenia in the eld-erly[OR(95％CI)=3.167(1.140-8.799),2.978(1.088-8.153),11.759(2.723-50.781),3.271(1.218-4.584)].High BMI,high albumin are protective factor for the occurrence of myopenia in the elderly[OR(95％CI)=0.691(0.570-0.838),0.737(0.642-0.845)].The H-L goodness of fit test results of the column chart prediction model showed that x2=7.039,P=0.532,the slope of the calibration curve approached 1,and the AUC in the ROC curve was 0.947(95％CI=0.919-0.975),in-dicating good consistency and discriminability of the predictive model in this column chart.There was a high net profit with-in the high-risk threshold range of 0.05-0.95,and the decision curve was above the All and None lines,and this column chart prediction model had high clinical practicality.Conclusion The age≥ 75 years old,osteoporosis,diabetes,high level of TG/HDL-C,low level of BMI,and low level of albumin are the influencing factors for the occurrence of sarcopenia in the elderly.The column chart model constructed by this method can effectively predict the occurrence of sarcopenia.