This report provides epidemiological figures about the characteristics of COVID-19 deaths during the first wave (1 March 2020 until 21 June 2020), the inter wave period (22 June until 30 August 2020) and the second wave (31 August 2020 until 14 February 2021) of the COVID-19 epidemic in Belgium. This is the period before the effects of the nationwide vaccination campaign that started early in 2021 could be assessed. In total 21,860 COVID-19 deaths occurred (43.9% in the first wave and 54.7% in the second wave). The COVID-19 mortality surveillance system was implemented at the start of the epidemic to acquire real-time COVID-19 mortality data on a daily basis. The surveillance combined information on COVID-19 related deaths from three surveillances (the hospital surveillance, the nursing home (NH) surveillance and notifications to regional health inspection authorities) through nine data sources. This information included the date of death, date of birth, sex, case classification, type of place of death, type of place of residence (e.g. living in a NH), postal code of the place of death and residence. Continuous improvements as regards the data collection resulted in retrograde adaptations of mortality numbers.

The overall sex distribution was fairly even (49.1% in male and 50.8% in female). Almost all deaths occurred in the age group over 64 years and approximately half of the deaths occurred in the age group over 84 years. Data on hospitalized COVID-19 patients showed that higher age, male sex and several comorbidities such as cardiovascular disease and diabetes were risk factors for mortality. Additionally, the estimated COVID-19 case fatality in Belgium confirmed that it was higher for the elderly and male population. In the second wave, more deaths occurred in hospitals (61%) than in nursing homes for elderly (NHs) (38%). In contrast, during the first wave, this distribution was more equal (50% in hospitals and 49% in NHs). The test capacity increased and the testing strategy broadened over time, leading to an increase in the proportion of laboratory-confirmed COVID-19 cases among deaths (69% and 95% in the first and second waves respectively). COVID-19 age-standardized mortality rates (ASMR), which take into account the age distribution of the population, showed that Brussels presented the highest ASMR for the total period and the first wave, while Wallonia has the highest ASMR for the second wave (more precisely in the provinces of Hainaut and Liège). The crude COVID-19 mortality rates for residents of NHs were higher in Flanders than in the other regions, both for the total period and for the second wave.

International comparison and ranking of COVID-19 crude mortality rates are misleading because of very heterogeneous methods used (e.g. case definition, testing and screening strategy, reporting method, availability of specific surveillance in NHs, etc.). Methods might also have changed during the course of the epidemic within the same country. A better comparison will probably be possible when countries have finished analyzing the official death certificates. The fast initiation of the COVID-19 surveillance in NHs and the inclusion of deaths of possible COVID-19 cases nevertheless allowed Belgium to provide accurate figures on COVID-19 deaths. This helped to assess the seriousness of the epidemiological situation in NHs. COVID-19 mortality was strongly correlated with excess all-cause mortality in Belgium. The excess mortality was a key indicator in the COVID-19 epidemic to validate that the epidemiological reporting of COVID-19-related mortality was correctly conducted during the epidemic