Inference of person-to-person transmission of COVID-19 reveals hidden super-spreading events during the early outbreak phase

Coronavirus disease 2019 (COVID-19) was first identified in late 2019 in Wuhan, Hubei Province, China and spread globally in months, sparking worldwide concern. However, it is unclear whether super-spreading events occurred during the early outbreak phase, as has been observed for other emerging viruses. Here, we analyse 208 publicly available SARS-CoV-2 genome sequences collected during the early outbreak phase. We combine phylogenetic analysis with Bayesian inference under an epidemiological model to trace person-to-person transmission. The dispersion parameter of the offspring distribution in the inferred transmission chain was estimated to be 0.23 (95% CI: 0.13–0.38), indicating there are individuals who directly infected a disproportionately large number of people. Our results showed that super-spreading events played an important role in the early stage of the COVID-19 outbreak. Although SARS-CoV-2 has spread rapidly, the contribution of super-spreading events to transmission is unclear. Here, the authors show that the number of secondary infections arising from an individual infection in the early phase of the outbreak was highly skewed, indicating that super-spreading events occurred.