Objectives: to describe the first wave of the COVID-19 pandemic with a focus on undetected cases and to evaluate different post-lockdown scenarios.
Design: the study introduces a SEIR compartmental model, taking into account the region-specific fraction of undetected cases, the effects of mobility restrictions, and the personal protective measures adopted, such as wearing a mask and washing hands frequently.
Setting and participants: the model is experimentally validated with data of all the Italian regions, some European countries, and the US.
Main outcome measures: the accuracy of the model results is measured through the mean absolute percentage error (MAPE) and Lewis criteria; fitting parameters are in good agreement with previous literature.
Results: the epidemic curves for different countries and the amount of undetected and asymptomatic cases are estimated, which are likely to represent the main source of infections in the near future. The model is applied to the Hubei case study, which is the first place to relax mobility restrictions. Results show different possible scenarios. Mobility and the adoption of personal protective measures greatly influence the dynamics of the infection, determining either a huge and rapid secondary epidemic peak or a more delayed and manageable one.
Conclusions: mathematical models can provide useful insights for healthcare decision makers to determine the best strategy in case of future outbreaks.
Keywords: COVID-19; SEIR; epidemiology lockdown.; mathematical models; public health.