Analysis of Risk Factors and Exploration of Predictors of Serious Cases of COVID-19 in Xi'an during the Period of 2021-2022

Autor: WANG Hai, WANG Zhuoli, PEI Honghong, PAN Longfei
Jazyk: čínština
Rok vydání: 2023
Předmět:
Zdroj: Zhongguo quanke yixue, Vol 26, Iss 17, Pp 2132-2137 (2023)
Druh dokumentu: article
ISSN: 1007-9572
DOI: 10.12114/j.issn.1007-9572.2022.0826
Popis: Background The outbreak of COVID-19 in Xi'an between 2021 and 2022 was a large-scale local epidemic in a large city with a huge number of cases. It is necessary to analyze and summarize the contents of this outbreak. Objective To analyze the disease characteristics of patients with COVID-19, and to explore the risk factors as well as predictors of serious cases. Methods General data and laboratory parameters were retrospectively collected from patients diagnosed with a new coronavirus pneumonia who were admitted to the Fourth People's Hospital of Xi'an between December 2021 and January 2022. Based on the the ratios of total IgG to lymphocyte percentage (IgG∶L%) , total IgM to lymphocyte percentage (IgM∶L%) , total IgG to lymphocyte count ratio (IgG∶L#) , and total IgM to lymphocyte count ratio (IgM∶L#) , patients were divided into three groups: mild and common, severe and critical. Multivariate Logistic regression analysis was used to explore the risk factors of developing severe and critically new coronavirus; then the ROC curve was drawn to analyze the predictive indexes and predictive value of severe and critical COVID-19, the area under the ROC curve (AUC) was calculated, and the AUC of each index was compared using the Delong test. Results A total of 699 patients with identified COVID-19 were finally included, and divided into two groups: the mild and common (n=678) and the severe and critical (n=21) forms, with the mild and common forms having younger age, and less underlying disease, D-dimer, IgM∶L%, IgM∶L#, and higher lymphocyte percentage and lymphocyte count than the severe and critical forms (P
Databáze: Directory of Open Access Journals