Transmissibility of coronavirus disease 2019 in Chinese cities with different dynamics of imported cases

BACKGROUND: Monitoring the reproduction number (R(t)) of the disease could help determine whether there is sustained transmission in a population, but areas with similar epidemic trends could have different transmission dynamics given the risk from imported cases varied across regions. In this study, we examined the R(t) of coronavirus disease 2019 (COVID-19) by taking different dynamics of imported cases into account and compared the transmissibility of COVID-19 at different intervention periods in Hangzhou and Shenzhen. METHODS: We obtained the daily aggregated counts of laboratory-confirmed imported and local cases of COVID-19 infections in Hangzhou and Shenzhen from January 1 to March 13, 2020. Daily R(t) and piecewise R(t) before and after Wuhan lockdown were estimated, accounting for imported cases. RESULTS: Since the epidemic of COVID-19 in Shenzhen was dominated by imported cases, R(t) was around 0.1 to 0.7 before the Wuhan lockdown. After the lockdown of Wuhan and the initialization of measures in response to the outbreak, local transmission was well-controlled as indicated by a low estimated value of piecewise R(t), 0.15 (95% CI [0.09–0.21]). On the contrary, R(t) obtained for Hangzhou ranged from 1.2 to 4.9 with a piecewise R(t) of 2.55 (95% CI [2.13–2.97]) before the lockdown of Wuhan due to the surge in local cases. Because of the Wuhan lockdown and other outbreak response measures, R(t) dropped below unity in mid-February. CONCLUSIONS: Even though Shenzhen had more cases than Hangzhou, local transmission did not sustain probably due to limited transmission from imported cases owing to the reduction in local susceptibles as residents left the city during Chunyun. The lockdown measures and local outbreak responses helped reduce the local transmissibility.