Re-estimation of basic reproduction number of COVID-19 based on the epidemic curve by symptom onset date

Previous studies have reported the basic reproduction number (R(0)) of coronavirus disease from publicly reported data that lack information such as onset of symptoms, presence of importations or known super-spreading events. Using data from the Republic of Korea, we illustrated how estimates of R(0) can be biased and provided improved estimates with more detailed data. We used COVID-19 contact trace system in Korea, which can provide symptom onset date and also serial intervals between contacted people. The total R(0) was estimated as 2.10 (95% confidence interval (CI) 1.84–2.42). Also, early transmission of COVID-19 differed by regional or social behaviours of the population. Regions affected by a specific church cluster, which showed a rapid and silent transmission under non-official religious meetings, had a higher R(0) of 2.40 (95% CI 2.08–2.77).