Impact of city and residential unit lockdowns on prevention and control of COVID-19

Peng Shao

The copyright holder for this preprint . https://doi.org/10.1101/2020.03.13.20035253 doi: medRxiv preprint Further, the TLU was set at 20 and the DLU at 1.0, 0.9, and 0.8, and the proportions of individuals in the five states and the number of available hospital beds were examined ( Figure 5) . A higher degree of implementation of measures to lock down residential units led to lower infection rates ( Figure 5A ; DLU = 1.0). At steady state, the proportion of S individuals was higher, and almost 85% of the individuals were not infected. This result showed that stringent implementation of measures to lock down residential unit could effectively control the spread of the epidemic. If the DLU was relaxed ( Figure 5C ; DLU = 0.8), the number of infected individuals significantly increased to 50% and the hospital bed shortages were prolonged. For example, under the complete lockdown of residential units ( Figure 5D ; DLU = 1.0), hospital bed shortage occurred at time frames of 20 -40. In contrast, hospital bed shortages occurred at time frames 20 -140 under non-stringent residential unit lockdown ( Figure 5F ; DLU = 0.8). The area under the curve (AUC) for the number of available beds that indicated the number of hospital bed shortages in a given period was smaller for complete lockdown ( Figure 5D ; DLU = 1.0) compared with the non-stringent lockdown of residential units ( Figure 5F ; DLU = 0.8). These findings indicated that more stringent implementation of the measures to lock down residential units led to lesser hospital bed requirements.