Household Transmission of SARS-CoV-2 Along the Evolution of Pandemic

The intra-household transmission of COVID-19 has recently received increasing attention due to its significant role in the spread of the pandemic. Here we developed a stochastic dynamic model motivated by the cluster growth algorithm in Erdös–Rényi Random Graph to adequately differentiate transmission within the household from that in the community due to the limited size of household members. This model allowed us to study the role and characteristics of household transmission within the full framework of virus transmission, not just the intrinsic characteristics of household transmission alone which was the main focus of existing studies. The model was applied to a comprehensive individual-level pandemic dataset collected in Yichang, China. We found that the household transmission accounted for 25.1% (12.9%-41.4%) and 38.5% (22.7%-54.4%) of the total infections before and during the lockdown. We assessed that 80.9% of infection was inevitable even if a containment measure was imposed to its extreme in Yichang from 24 January 2020 that cut all routes of community transmission. Our model also suggested that the household-level contact tracing could have reduced the number of infections by more than 50% and advanced the clearance date of active infection by 72 days. In addition, the vaccination programs would indeed reduce the total infection significantly but in the meanwhile might potentially prolong the pandemic by increasing the covertness of the disease.Funding: This research is supported by National Natural Science Foundation of China grant 8204100362 and the Bill & Melinda Gates Foundation (INV-016826).Declaration of Interest: There is no conflict of interestEthical Approval: This study was approved by the ethics committee of the Yichang CDC, de-identified data was used for the analysis.