Modeling Exit Strategies from COVID-19 Lockdown with a Focus on Antibody Tests

Reinhard German; Anatoli Djanatliev; Lisa Maile; Peter Bazan; Holger Hackstein

The importance of AB tests would be even greater if the disease spreads more drastically, e.g., due to non-optimal interventions. This in turn would lead to more infections and in turn to more asymptomatic recovered and thus increase the likelihood of finding people with antibodies, making the tests more effective and excluding more people from interventions. However, we have not added figures for these scenarios because the ICU capacities would be exceeded and the number of deaths would increase significantly, as discussed in the previous scenarios. Fig. 10 and 11 demonstrate how the virus spreading would be affected if a seasonality factor is in place, meaning that the virus spreads less in summer and more in winter months, or if a significant portion of the population is immune to the virus. The seasonality is modeled as sine curve, as also explained in the Appendix. The figures show the dynamics with no antibody testing in place, the number of immune people and thus the contact index could be improved with the previously described measure by the number of unknown immune people tracked by the bottom right curve. We omitted the graphics for space reasons but the number of additionally identified immune people behaves nearly identically as with the effect shown in Fig. 8 and 9 . As we can see, the dynamics of the virus are similar, the disease will still spread fast after April 19 and would exceed ICU capacities. The main difference is the number of interventions that will be required, which reduces from 9 to 6 (for seasonality) or 5 (for initially immune people). If an initial immunity of one third of the population is assumed, the last intervention is already in November 2021. To optimally propose interventions, our model can thus be adapted when more information for this behavior is known. However, currently, we assume the worst-case behavior without seasonality and no initial immunization.