Modelling the SARS-CoV-2 first epidemic wave in Greece: social contact patterns for impact assessment and an exit strategy from social distancing measures

In Greece, a nationwide lockdown to mitigate the transmission of SARS-CoV-2 was imposed on March 23, 2020. As by the end of April the first epidemic wave is waning, it is important to assess the infection attack rate and quantify the impact of physical distancing. We implemented a survey to assess social mixing patterns before the epidemic and during lockdown. We estimated R ₀ from surveillance data and assessed its decline as a result of physical distancing based on social contacts data. We applied a Susceptible-Exposed-Infectious-Recovered model to estimate the infection attack rate and the infection fatality ratio (IFR). As multiple social distancing measures were implemented simultaneously (schools/work/leisure), we assessed their overall impact as well as their relative contribution. R ₀ was estimated 2·38 (95%CI: 2·01,2·80). By April 26 ^th , the infection attack rate was 0·12% (95%CrI: 0·06%,0·26%) and the IFR 1·12% (95%CrI: 0·55%,2·31%). During lockdown, daily contacts were reduced by 86·9% and the effective reproduction number reached 0·46 (95%CrI: 0·35,0·57). The reduction in R ₀ attributed to lockdown was 81·0% (95%CrI: 71·8%,86·0%) whereas the reduction attributed to each measure separately ranged between 10%-24%. We assessed scenarios with less disruptive social distancing measures as well as scenarios where measures are partially lifted after lockdown. This is the first impact assessment of the first wave of SARS-CoV-2 in a European country. It suggests that only multiple measures implemented simultaneously could reduce R ₀ below 1. Measuring social mixing patterns can be a tool for real-time monitoring of the epidemic potential.