Реферат
Objective To analyze the risk factors of repeated admissions for community-acquired pneumonia(CAP)in adults,and to build a nomogram model to predict individual risk.Methods A total of 2 306 adult hospitalized patients with CAP in Beijing Shijitan Hospital affiliated to Capital Medical University from January 2018 to Decem-ber 2020 were retrospectively selected and divided into repeated admission group and control group according to whether they were readmitted within one year after discharge.Univariate logistic analysis and multivariate logistic a-nalysis were used to determine the risk factors.The risk factors were introduced into R 3.5.3 software to construct the nomogram prediction model.The calibration curve was drawn and the Hosmer-Lemeshow goodness of fit test was performed to evaluate the accuracy of the nomogram prediction model.The receiver operating characteristic(ROC)curve was drawn to evaluate the discrimination of the nomogram prediction model.The decision analysis curve was drawn to measure patient benefits.Results The age,gender,length of stay,total score of comorbidity index,use of special grade antibiotics,history of blood transfusion,and Vaccination history were risk factors of repeated hospi-tal admissions for CAP in adults.The nomogram prediction model of adult CAP repeated admission was constructed based on the above risk factors.The Hosmer-Lemeshow goodness of fit test showed that the fitting effect of the no-mogram prediction model was good(x2=8.873,P=0.353).The ROC curve analysis showed that the area under the curve of the test dataset was 0.775.The results of the decision analysis curve showed that when the threshold was 0.21,the model could generate a net profit of 0.104.Conclusion The nomogram model established in this study has good discrimination and accuracy in predicting the risk of adult CAP repeated admissions and has high clinical application value.