LAUSR

Purpose The incubation period is an important epidemiologic characteristic of infectious diseases in determining the quarantine period. In South Korea, there still have been debates about the quarantine period of coronavirus 2019 (COVID-19). Furthermore, the differences in the incubation period of COVID-19 by age and gender are still not well understood. Methods & Materials We collected data on COVID-19 cases published by the South Korean public health authorities. Using this data, we estimated the incubation period by fitting three different distributions (Weibull, gamma, and log-normal) by gender, age group, and the different epidemic periods of COVID-19. We divided our study into two epidemic periods (First epidemic wave: 28 January 2020 – 18 April 2020, Second epidemic wave: 19 April 2020 – 30 August 2020). We used the Wilcoxon test to assess for any significant differences between the incubation periods by gender, epidemic period, and age group. We selected the best-fit model by comparing the Akaike Information Criterion. All analyses were done in R version 3.6.1 and level of significance was set at p-value < 0.05. Results The log-normal model was best fitted in the study. The estimated median incubation period using the log-normal model was 4.6 days (95% confidence interval: 1.19 - 1.34), and the 95th percentile was 11.74 days. There was no significant difference in incubation period between males and females (P = 0.42), as well as with the epidemic periods (P = 0.77). Conclusion This study provides evidence for the median incubation period for COVID-19 of approximately 4.6 days. Our work brings out more evidence of the incubation period for COVID-19 and shows that it may be prudent to continue with the current 14-day quarantine policy.