Risk factors for tuberculosis in AIDS patients and establisninent of prediction model
OBJECTIVE To investigate the epidemiology of AIDS patients with tuberculosis and construct a predic-tion model.METHODS Totally 2 736 patients with acquired immune deficiency syndrome admitted to Hangzhou Xixi Hospital from Jan.2019 to Dec.2021 were included in the study,of which 72 patients with acquired immune deficiency syndrome combined with tuberculosis were included in the acquired immune deficiency syndrome com-bined with tuberculosis group and 2664 patients with acquired immune deficiency syndrome were included in the acquired immune deficiency syndrome simple group,and the clinical data of patients were retrospectively collected.The status of acquired immune deficiency syndrome and tuberculosis,information on diagnosis and treatment of acquired immune deficiency syndrome and tuberculosis were analyzed,and the risk factors of AIDS and tuberculo-sis were analyzed by Logistic regression analysis.A nomogram prediction model was constructed,and its predic-tive value for AIDS with tuberculosis was analyzed by receiver operating characteristic(ROC)curve.RESULTS From 2019 to 2021,the incidence of newly diagnosed AIDS with tuberculosis was 2.63％,and there was no statis-tical difference in the incidence rate of AIDS with tuberculosis in each year.In 2019,2020 and 2021,there was no statistically significant difference in the infection site,history of tuberculosis,time from symptoms to treatment,and anti tuberculosis treatment of AIDS patients with tuberculosis.Logistic regression analysis showed that intra-venous drug use,education level,CD4+cell count,body mass index(BMI)were the risk factors of AIDS with tu-berculosis(P＜0.05).The nomogram prediction model of risk factors of AIDS combined with tuberculosis was constructed.The results of Hosmer Lemeshow fit test showed that the accuracy of model fitting was good.Logis-tic regression model was used to statistically analyze the data,and the diagnostic probability of the risk model of AIDS combined with tuberculosis was logit(P)=-0.920+AIDS infection route × 0.977+education level × 0.554+CD4+cell number × 1.017-BMI × 0.209.The ROC curve of AIDS complicated with tuberculosis was drawn according to the diagnostic probability logit(P).When logit(P)＞0.026,the area under the curve(AUC)value was 0.753,indicating that the model was well differentiated.CONCLUSION AIDS-combined tuberculosis was strongly associated with a variety of factors,and the construction of model could be used for early screening of high-risk patients and preventive intervention.
Acquired immune deficiency syndromeTuberculosisStatusRisk factorPrediction model
NETL
NSTL
万方数据
