Figure 3.

Secondary outbreak appearing after the home-based confinement is removed. The lockdown is removed and the parameter values are as during the week of March 16th−22nd. We simulated different degrees of quarantines to explore its effect on the second wave. The greater the percentage of infected cases that are quarantined at symptom onset, the smaller the incidence of the second wave. The shaded area represents the 95% CI (for clarity, only shown for the case of no quarantine). Confidence intervals calculated using Cox's method (26). (A) First infected case is on February 7th, the number of infected in the first wave has not reached herd immunization and a second wave is triggered by a single infected case. The intensity of the second wave is one order of magnitude larger than the first. (B) First infected case is on January 28th, the number of infected cases in the first wave has reached a larger fraction of the population and the intensity of the second wave is, in the scenario, smaller than the first. Average over 100 realizations.