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

Peng Shao

Computational experiments were conducted in two-dimensional space with a population size of NU. In the initial phase, NU was randomly and uniformly distributed in NC; that is, the number of individuals in each city was NU/NC. One individual in a major city was randomly selected as the first infected person. The simulation program ran 200 timeframes, during which individuals might undergo spatial movement and a state change at any time frame (for example, daily). First, without considering the status of city and residential unit lockdowns, individuals moved to neighboring cities at a probability of the across-community mobility rate of the population (MU). Secondly, individuals in all cities either changed or maintained their original states based on the rules of the SAIRD model. Figure 2 shows differences in the ratios of individuals in the five states in the core locked down cities and the ratios of infected users in each city. With respect to timing, if a city was locked down after an infected user was confirmed and given no residential unit lockdown and no increase in medical resources, the ratio of individuals in S state would be 0% at steady state; that is, all individuals would be infected. This was without regard to whether the city was under early ( Figure 2C ), late ( Figure 2B ), or no lockdown (Figure 2A ). Regarding variance in the ratios of D individuals among all cities, the earlier a city was locked down ( Figure 2F , TLC = 5), the greater the variance in the proportions of D individuals state across all 81 cities (VARD ~ 0.05). This indicated that the number of deaths in the core cities that were under lockdown was significantly higher than in other cities. By comparison, variance in the proportions of D individuals across the 81 cities was lower when no cities were locked down (VARD of ~ 0.02; Figure 2D ). Overall, the implementation of measures to place cities under lockdown could not reduce the ratios of infected individuals in the SAIRD model. The implementation of measures to lock down cities would result in larger variances in the number of deaths between cities with and without lockdown. These findings were consistent with the perspective of Li et al. (2020) , who reported that the stringent lockdown policy applied to Hubei Province had suspended a nationwide outbreak of COVID-19 [8] . However, mortality rates in lockdown cities were higher because infected individuals in core cities were unable to move to other cities. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.