COVID-19 mortality, excess mortality, deaths per million and infection fatality ratio, Belgium, 9 March 2020 to 28 June 2020

Abstract

Background: COVID-19 mortality, excess mortality, deaths per million population (DPM), infection fatality ratio (IFR) and case fatality ratio (CFR) are reported and compared for many countries globally. These measures may appear objective, however, they should be interpreted with caution. Aim: We examined reported COVID-19-related mortality in Belgium from 9 March 2020 to 28 June 2020, placing it against the background of excess mortality and compared the DPM and IFR between countries and within subgroups. Methods: The relation between COVID-19-related mortality and excess mortality was evaluated by comparing COVID-19 mortality and the difference between observed and weekly average predictions of all-cause mortality. DPM were evaluated using demographic data of the Belgian population. The number of infections was estimated by a stochastic compartmental model. The IFR was estimated using a delay distribution between infection and death. Results: In the study period, 9,621 COVID-19-related deaths were reported, which is close to the excess mortality estimated using weekly averages (8,985 deaths). This translates to 837 DPM and an IFR of 1.5% in the general population. Both DPM and IFR increase with age and are substantially larger in the nursing home population. Discussion: During the first pandemic wave, Belgium had no discrepancy between COVID-19-related mortality and excess mortality. In light of this close agreement, it is useful to consider the DPM and IFR, which are both age, sex, and nursing home population-dependent. Comparison of COVID-19 mortality between countries should rather be based on excess mortality than on COVID-19-related mortality. ; The seroprevalence study of which the results are used in this manuscript has been sponsored by the University of Antwerp's Research Fund. This project has received funding from the European Union's Horizon 2020 Research and Innovation Programma -Project EpiPose (No 101003688). SA, LW and NH gratefully acknowledge support from the Fonds voor Wetenschappelijk Onderzoek (FWO) (RESTORE project -G0G2920N and postdoctoral fellowships 1234620N). The investigators were independent from the funders.