Transmission Dynamics and the Effects of Non-Pharmaceutical Interventions in the COVID-19 Outbreak Resurged in Beijing, China: a Descriptive and Modelling Study

OBJECTIVE:To evaluate epidemiological characteristics and transmission dynamics of COVID-19 outbreak resurged in Beijing and to assess the effects of three non-pharmaceutical interventions.DESIGN:Descriptive and modelling study based on surveillance data of COVID-19 in Beijing.SETTING:Outbreak in Beijing.PARTICIPANTS:The database included 335 confirmed cases of COVID-19.METHODS:To conduct spatiotemporal analyses of the outbreak, we collected individual records on laboratory-confirmed cases of COVID-19 from 11 June 2020 to 5 July 2020 in Beijing, and visitor flow and products transportation data of Xinfadi Wholesale Market. We also built a modified susceptible-exposed-infected-removed model to investigate the effect of interventions deployed in Beijing.RESULTS:We found that the staff working in the market (52.2%) and the people around 10 km to this epicentre (72.5%) were most affected, and the population mobility entering-exiting Xinfadi Wholesale Market significantly contributed to the spread of COVID-19 (p=0.021), but goods flow of the market had little impact on the virus spread (p=0.184). The prompt identification of Xinfadi Wholesale Market as the infection source could have avoided a total of 25 708 (95% CI 13 657 to 40 625) cases if unnoticed transmission lasted for a month. Based on the model, we found that active screening on targeted population by nucleic acid testing alone had the most significant effect.CONCLUSIONS:The non-pharmaceutical interventions deployed in Beijing, including localised lockdown, close-contact tracing and community-based testing, were proved to be effective enough to contain the outbreak. Beijing has achieved an optimal balance between epidemic containment and economic protection.