Analysis Of Factors Associated With Severe Illness Of Novel Coronavirus Infection And Construction Of A Diagnostic System

Object:The novel coronavirus infection has been a worldwide pandemic for over three years and although the number of cases is now significantly lower,it still shows no signs of disappearing.The aim of this study is to construct a diagnostic system for people at high risk of severe neocoronavirus infection and to screen people at high risk of severe disease and serious illness through dynamic assessment of their condition,so as to provide reference for early diagnosis,rational treatment and rational use of medical resources,and to provide reference for effective prevention and reduction of severe neocoronavirus infection.Method:Based on a retrospective single-centre case series design,data from 124 cases of novel coronavirus infection(of the original strain)admitted from January to July 2020and 644 cases of novel coronavirus infection(of the Delta and Omicron variants,collectively referred to as subsequent strains)admitted from September 2021 to August2022 at the Chengdu Public Health Clinical Medical Centre were used to describe case characteristics and explore risk factors for the occurrence of severe disease through univariate analysis and logistic regression analysis,and the predictive effect of the model was evaluated using subject operating characteristic(ROC)curves.Results:1.Data from patients with the original strain of novel coronavirus infection were analysed and the results of the univariate analysis showed that age,combined underlying disease,days in hospital,neutrophils,c-reactive protein,portal aminotransferase of liver function,myoglobin,high-sensitivity troponin T,N-terminal brain natriuretic peptide,prothrombin time,urinary fibrin degradation products,D-dimer,CD4~+and CD8~+were risk factors for patients who were likely to develop severe disease.After multiple linear regression and combined with clinical practice,combined underlying disease,age,c-reactive protein level and CD4~+count level were used to build the diagnostic system and again included in logistic regression for model fitting,the area under the ROC curve for the scoring model was obtained as 0.984(95%CI:0.968 to 1.000)with a best cut-off value of 13.588(sensitivity of 0.976 and a specificity of 0.916).2.Data from patients with a subsequent strain of novel coronavirus infection were analysed and the univariate analysis showed that gender,age,number of vaccinations,comorbid underlying diseases,epidemiological history,c-reactive protein,and CD4~+were risk factors for the possibility of severe illness in patients.The four key indicators screened(combined underlying disease,age,c-reactive protein and CD4~+count)were included in the logistic regression for model fitting and the area under the ROC curve for the scoring model was obtained as 0.969(95%CI:0.955 to 0.983)with a best cut-off value of 9.07(sensitivity of 0.982 and specificity of 0.852).3.Data from patients infected with the original strain of neocoronavirus were substituted into the constructed diagnostic system for critically ill patients infected with subsequent strains of neocoronavirus for validation,and the results of plotting the ROC curve showed an AUC of 0.97(95%CI:0.95-0.99,P<0.001),sensitivity of 0.951 and specificity of 0.880.4.When comparing the factors affecting patients with severe neocoronavirus infection with those with severe subsequent infection,we found that the proportion of patients with severe neocoronavirus infection with hypertension,diabetes and cardiovascular disease was higher than that of patients with severe subsequent infection,the median c-reactive protein,days in hospital and age were higher in patients with severe neocoronavirus infection than in patients with severe subsequent infection,and the median CD4~+count was lower in patients with severe neocoronavirus infection than in patients with severe subsequent infection.Conclusion:The diagnostic system established in this study for patients with severe neocoronavirus infection has good discriminant validity.The system can identify the severity of the disease,determine the prognosis at an early stage,and assess the patient’s condition to provide some reference for subsequent treatment.The observation of changes in the scoring system is beneficial to the dynamic management of patients with neocoronavirus infection.