Risk Factor Analysis And Risk Prediction Model Of Adverse Treatment Outcome In Patients With Pulmonary Tuberculosis

Objective:To explore the relevant risk factors of adverse treatment outcome in patients with pulmonary tuberculosis,build a model to predict the risk of adverse treatment outcome in patients with pulmonary tuberculosis,so as to provide theoretical basis for effective treatment of pulmonary tuberculosis,to provide technical support for screening high-risk population with adverse treatment outcome.Methods:The medical records of tuberculosis patients admitted to the Department of Infection of the First Affiliated Hospital of Shihezi University from May 2016 to May 2021 were retrospectively collected.Each medical record was reviewed one by one and the general demographic data(gender,age,nationality,occupation,residence)and clinical data(complications,smoking history,drinking history,sputum smear,blood routine,blood biochemistry,lung CT,treatment type,discharge diagnosis)were collected,The treatment outcome of each tuberculosis patient was obtained by telephone follow-up one by one,and the risk factors of adverse treatment outcome of tuberculosis were analyzed by multivariate unconditional logistic regression.All cases were randomly divided into the modeling group and the validation group according to the ratio of 7:3.The risk prediction model was established by logistic regression,the goodness of fit of the model was evaluated by using the Hosmer-Lemeshow test model,the area under the ROC curve(AUC)was drawn to evaluate the discrimination ability of the model,and the maximum value of the Joden index was selected as the best cutoff point to perform the back generation validation of the model.Results:1.A total of 930 cases of pulmonary tuberculosis were included in this study.The results of pulmonary tuberculosis treatment were as follows: 430 cases were cured,249 cases completed the course of treatment,65 cases failed,80 cases died of tuberculosis,106 cases died of non-tuberculosis,the unsuccessful rate of treatment was 26.99%,and the mortality rate was 20%.2.Logistic regression analysis showed that,male,age ≥60 years old,respiratory failure,chronic obstructive pulmonary disease,malignant tumor,Hypoalbuminemia,anemia and re-treatment were the risk factors of unsuccessful treatment The risk factors of death in patients with tuberculosis were male,age ≥60 years,anemia,respiratory failure,chronic obstructive pulmonary disease,AIDS,retreatment and Hypoalbuminemia.3.The AUC of the risk prediction model was 0.815,the result of H-L test was P>0.05,and the probability value above 0.18 in the validation group was judged as high risk,and the probability value below 0.18 was judged as low risk,the sensitivity and specificity of the model were 0.845 and 0.625,respectively.The positive predictive value and negative predictive value of the model were 43.48% and 92.20%,respectively.Conclusion:1.The incidence of adverse treatment outcome of pulmonary tuberculosis is high.The relevant departments should pay more attention to the treatment effect of pulmonary tuberculosis patients.2.Male sex,age ≥60 years,respiratory failure,anemia,Hypoalbuminemia,retreatment,chronic obstructive pulmonary disease,malignant tumor,AIDS are risk factors for adverse treatment outcomes in patients with tuberculosis,relevant departments should take targeted measures to achieve the focus of prevention and control of the relevant populations.3.The risk prediction model for adverse treatment outcome of pulmonary tuberculosis patients based on the case history data of pulmonary tuberculosis patients has a good overall prediction ability,which can provide a basis for the prediction of high-risk patients,and has a certain application value.