Epidemiological parameter estimation and characteristics of the novel coronavirus disease (COVID-19) transmission in Shaanxi Province / 公共卫生与预防医学

ABSTRACT

Objective To estimate the key epidemiological parameters, including the serial interval (SI), basic reproduction numbers (R0), and time-dependent reproduction numbers (Rt) based on the case data of COVID-19 published on the official website of Shaanxi Provincial Health Commission. Methods The method of maximum likelihood estimation was used to fit the probability distribution of SI, and three parametric models including Gamma, Weibull, and Lognormal distributions were adopted to estimate the distribution of SI. The optimal model was selected by the corrected Akaike information criterion (AICc). Based on nonlinear regression model the cumulative number of confirmed cases was fit to the growth rate r of the Richard growth model to estimate R0. The Rt was calculated based on the Bayesian framework. Results A total of 49 transmission chains were discovered from 245 local confirmed cases in Shaanxi Province. The Gamma distribution was the optimal model to fit the SI here by AICc. Using the Gamma distribution, the mean SI was estimated to be 6.3 days (95%CI5.98 - 6.43) with a standard deviation (SD) of 3.94 days (95%CI 3.01 - 5.03). Using the Richard growth model, the growth rate was estimated to be 0.23 (95% CI 0.21 - 0.24) and the basic reproduction number in Shaanxi to be 3.11 (95%CI 2.91 - 3.40). Rt showed an overall downward trend, and fell below 1 on February 10 (Rt=0.95（95%CI 0.76-1.16)), and stabilized at around 0.35 on February 18. Conclusion The SI of COVID-19 is relatively shorter than that of MERS and SARS, while the R0 is relatively larger, and Rt is on a downward trend, which suggests COVID-19 is a highly transmissible infectious disease. The control measures including the isolation and treatment of confirmed patients, quarantine and observation of suspected cases, contact tracing, improvement of public awareness, and adoption of self-protection measures can effectively reduce the COVID-19 outbreak.