Fig 1.
Number of patients suspected to be infected with SARS-CoV-2 directed to the triage centers located close to the emerging departments of the Liège University Hospital (N = 4,489) in function of weeks as well as those that were tested by qRT-PCR and included in the present study (N = 2,152).
Patients suspected to be infected with SARS-CoV-2 (black circles) and patients tested by qRT-PCR and included in the study (white circles).
Table 1.
Demographic characteristics and clinical symptoms of the study population of patients (N = 2,152).
Table 2.
Odds ratio with 95% confidence interval (CI) derived from univariate binary logistic regression, for two demographic characteristics (age and gender) and eleven symptoms in 573 confirmed and 1579 unconfirmed COVID-19 patients.
Table 3.
Odds ratio with 95% Confidence Interval (CI) derived from multivariate binary logistic regression, calculated for two demographic characteristics (age and gender) and symptoms which were found to be significant at the univariate level, when considering p-value < 0.2, in 573 confirmed and 1579 unconfirmed COVID-19 patients.
Table 4.
Matrix of binary Jaccard similarity coefficients between clinical signs and the COVID-19 status.
Fig 2.
Receiver operating characteristic curve of the overall pondered score of COVID-19.
Points are the observed values; the solid curve in black and its 95% confidence interval (broken curves in black) was fitted according to a binormal distribution. Area under curve = 0.71 (95% CI: 0.68–0.73) with standard error = 0.012.
Fig 3.
The evolving of the Overall Pondered Score (OPS) in function of the period of the first COVID-19 pandemic wave.
Table 5.
The probability for a patient to be confirmed and unconfirmed COVID-19 cases in function of the result of the Overall Pondered Score (OPS).
Fig 4.
Classification decision tree for clinically suspected COVID-19 cases at the Liège University Hospital (N = 2,152).
(1) and (0), presence and absence of the symptom; Class, in blue or red, the number of confirmed or unconfirmed patients to SARS-CoV-2, respectively.
Table 6.
Area under the receiver operating characteristic curve (AUC-ROC) in function of the data set used.