Establishment of and empirical research on clinical stratification and early warning model for the risk of upper gastrointestinal bleeding in the elderly
WEI Jinglin
SUN Jinhua
LIU Lianpei
LI Zhonghao
ZHAI Zhenzhen
WANG Hongbin
Abstract：Objective To establish a clinical stratification and early warning model for the risk of upper gastrointestinal bleeding(UGIB)in the elderly,and to conduct an empirical analysis.Methods A total of 200 elderly patients with UGIB admitted from January 2023 to January 2024 were selected and divided into high-risk group(n=91)and non-high-risk group(n=109)according to the risk degree.The clinical data of the two groups at admission were compared,and multivariate logistic regression was used to analyze the related influencing factors of high-risk UGIB in the elderly.A clinical stratification and early warning model was established and a nomogram was drawn.In addition,internal and external data were used to verify the predictive value of this model for high-risk UGIB in the elderly.Results Hepatic,renal and cardiac failure,previous UGIB,shock index,hemoglobin,prognostic nutritional index(PNI),international normalized ratio(INR),red blood cell distribution width to lymphocyte ratio(RLR)and gastric cancer were related influencing factors for high-risk UGIB in the elderly(P<0.05).The consistency index of the clinical stratification and early warning model for the risk constructed by the above related influencing factors as independent variables was 0.843,and the area under the receiver operating characteristic curve was 0.923(95%CI:0.876,0.956).The consistency between the results of the clinical stratification and early warning model for the risk and the clinical reality of UGIB in the elderly was 96.00%,and the Kappa value was 0.918(95%CI:0.641,1.195).Conclusion Hepatic,renal and cardiac failure,previous UGIB,shock index,hemoglobin,PNI,INR,RLR and gastric cancer are all related influencing factors for high-risk UGIB in the elderly.The clinical stratification and early warning model constructed on this basis has high discrimination and accuracy,and can effectively screen high-risk groups of UGIB in the elderly.
Keywords：upper gastrointestinal bleedingelderlyrisk stratificationlogistic regression modelearly warning modelreceiver operating characteristic
Publication Date：2025-08-28
Online Publishing Date：2025-09-12(First online date of this platform, not the publication date of the document)
Pages：6( 25-30 )