Realistic estimate of the Covid-19 incidence and mortality rate in France

TL;DR

This study estimates the true Covid-19 incidence and mortality in France using large-scale testing data, revealing a much higher spread than previously thought and a significant reduction in severity since the first wave.

Contribution

It provides a realistic estimate of Covid-19 incidence and mortality in France and models outbreak evolution considering herd immunity effects.

Findings

Incidence at 0.8% in August rising to 2.4% in September

ICU admission rate and IFR dropped by one order of magnitude since March

Herd immunity likely achieved by end of October 2020

Abstract

Large scale virological testing of SARS-Cov2 is implemented since May 2020 in France. We assume that the positivity of asymptomatic people not being contact cases (ANBC) is representative of the positivity of the whole French population, which allows estimating the real incidence. We estimate, using Sant\'e Public France reports, that the incidence at the beginning of August was about 0.8% and rose to about 2.4% of the total population at the beginning of September. This corresponds to about 1.6 million people simultaneously infected and 230.000 new infections each day. These evaluations allow to deduce that intensive care units (ICU) admission rate and infection fatality rate (IFR) dropped by one order of magnitude since March, and are currently 0.036% and 0.027% respectively. Basic simulations of the outbreak evolution based on the hypothesis of negligible reinfection probability are performed for France, Ile de France, Puy de Dome, Bouches du Rhone and Grand Est. These simulations are using a reproduction rate (R) constant over time ranging from 1.3 to 1.45 (1.15 in Grand Est). They use an estimation of the number of infection which occurred during the first wave. These simulations also use the estimated incidence, which by reducing the susceptible population weeks after weeks induces a saturation of the otherwise exponential growth of the outbreak. An incidence peak of 3.5 % is expected at week 39 for France. The calculated total number of ICU admission and deaths during the second wave are 9000 and 7000 respectively for R=1.3. The cumulative incidence over the two waves is computed close to 60% for R=1.3 and 70% for R=1.4. This suggests that if individual immunity exists, herd immunity is likely to be achieved in France by the end of October 2020. We conclude that Covid-19 is much more spread than previously thought, but its severity became limited since the end of the first wave.