Statistical Analysis

We constructed epidemic curves using the date of illness onset and the date of COVID-19 confirmation. Furthermore, we estimated the time delay between the date of illness onset and the date of COVID-19 confirmation by fitting log-normal, gamma, and Weibull distributions using the available data [3]. We identified probabilistic distribution models and evaluated the suitability of the distribution models through the Bayesian information criterion (BIC) prior to the statistical analysis [4,5]. All statistical analyses were performed using R version 3.0.2 (https://cran.r-project.org/) and the R package “MASS” was used to identify the best model fit.