Estimated number of deaths directly averted in people 60 years and older as a result of COVID-19 vaccination in the WHO European Region, December 2020 to November 2021

Since December 2019, over 1.5 million SARS-CoV-2-related fatalities have been recorded in the World Health Organization European Region - 90.2% in people ≥ 60 years. We calculated lives saved in this age group by COVID-19 vaccination in 33 countries from December 2020 to November 2021, using weekly reported deaths and vaccination coverage. We estimated that vaccination averted 469,186 deaths (51% of 911,302 expected deaths; sensitivity range: 129,851–733,744; 23–62%). Impact by country ranged 6–93%, largest when implementation was early.

Since December 2019, over 1.5 million SARS-CoV-2related fatalities have been recorded in the World Health Organization European Region -90.2% in people ≥ 60 years. We calculated lives saved in this age group by COVID-19 vaccination in 33 countries from December 2020 to November 2021, using weekly reported deaths and vaccination coverage. We estimated that vaccination averted 469,186 deaths (51% of 911,302 expected deaths; sensitivity range: 129,851-733,744; 23-62%). Impact by country ranged 6-93%, largest when implementation was early.
Since the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was first detected in December 2019 [1], in excess of 1.5 million coronavirus disease (COVID- 19) fatalities have been reported in the World Health Organization (WHO) European Region until week 45/2021, with 90.2% deaths in people 60 years and older [2]. In a subset of 33 countries in the Region covered by this study, 442,116 fatalities were reported between December 2020 and November 2021 in this age group. We observed a wide range in cumulative national mortality rate from 5.4 to 1,008.0 per 100,000 population 60 years and older. Here, we estimate the number of deaths averted in this age group since the start of COVID-19 vaccination in those countries of the WHO European Region with available data.

Vaccination roll-out and uptake
The emergence of SARS-CoV-2 was followed by the rapid development, licensure and roll-out of several COVID-19 vaccines from late 2020 onwards, initially targeting select groups including those at higher risk of severe disease, in particular older adults. Fiftyone countries, areas or territories in the Region had reported the weekly number of vaccination doses administered to The European Surveillance System (TESSy) until week 45/2021, of which 33 countries had reported age-specific mortality counts and number of vaccination doses administered. Vaccination uptake (VU) in these countries increased in priority groups such that by week 45, 80% (range: 20-100) of people 60 years and older had received a complete vaccination series and 84% (range: 29-100) had received at least one dose (Table 1, Figures 1, 2, 3). The impact of such vaccination campaigns on preventing severe disease in terms of averted deaths can be quantified [3,4], although an assessment of lives saved using a standardised methodology across multiple countries in Europe is lacking.

Number of deaths averted as a result of vaccination among older adults
Data from 33 countries reporting both COVID-19 agespecific vaccination and age-aggregated COVID-19 mortality data for people 60 years and older were downloaded from TESSy on 18 November 2021 and restricted to the weeks between 51/2020 and 45/2021. Thirty countries reported more granular age groups (60-69, 70-79 and ≥ 80 years). Population estimates for each age group in year 2020 were drawn from the United Nations [5] for countries not in the European Union (EU) or European Economic Area (EEA) and from Eurostat [6] for EU/EEA countries, except for Israel (age denominators from Israel Central Bureau of Statistics [7]). All population data were downloaded in 2021. All analyses presented here were conducted in R software version 4.0.5 (R Foundation, Vienna, Austria) [8]. We assumed that all countries used the same definition for COVID-19 mortality and that reporting delays were comparable. Week Vaccination uptake %

Vaccine dosage
First dose Full coverage WHO: World Health Organization.
The weekly number of deaths averted per country was estimated using methods adapted from Machado et al. [9] (available on GitHub [10]), with the following definitions: VE 1 as vaccine effectiveness (VE) (at least one dose), VE 2 as vaccine effectiveness (complete series), VU 1 being vaccination uptake (at least one dose) and VU 2 as vaccination uptake (complete series). The rolling average number of deaths observed over three consecutive weeks centred on the second week was used for the calculations (the true number of weekly fatalities is shown in Tables 1 and 2).
The weekly number of deaths averted for each dose was calculated as below for each age group, setting the base scenario for VE against COVID-19 mortality at 60% and 95% for VE 1 and VE 2 , respectively [11][12][13]. For single-dose vaccines, the VE of the complete vaccination series was equal to the first dose.
Here, w represents the time delays of 2 and 1 weeks for the development of a full immune response after, respectively, the first and second vaccine dose [14] and the median time from infection to death of 2 weeks [15,16]. The number of deaths averted each week was then added to the number of observed deaths to calculate the total expected number of deaths and cumulative mortality rate per 100,000 population per country. Using the base VE scenario, we calculated that 51% (n=469,186) of total expected deaths (n=911,302) were averted by vaccination over the study period; ranging from 93% of deaths averted in Iceland to 6% in Ukraine ( Table 1). All three countries with 60% or less of their population 60 years and older fully vaccinated by week 45/2021 (Moldova, Romania and Ukraine; Table 1) had a maximum of only 20% expected deaths averted over the study period ( Figure 3). On the contrary, all four countries (Israel, Malta, United Kingdom (UK)-England and UK-Scotland) that achieved very high complete vaccination coverage (above 90%) already by week 23/2021 ( Figure 1) averted over 65% of total expected deaths by week 45 (Table 1 and Figure 3). In the 30 countries with more detailed data on age groups allowing for inclusion in age-stratified analyses, the largest number of fatalities were averted after vaccination among people 80 years and older (261,421 fatalities, 57% of total averted deaths). A total of 140,694 (31%) fatalities were averted among 70-79 year-olds and 55,966 (12% of averted deaths) among 60-69 year-olds in the base scenario (Table 2).

Sensitivity analyses investigating assumed immune response delays and vaccine effectiveness
To investigate the parameters used in the base scenario using VE 1 (60%) and VE 2 (95%), we ran sensitivity analyses varying the time to protection and time from infection to death using shorter (3 and 2 weeks, respectively) and longer (5 and 4 weeks, respectively) time ranges.
A sensitivity analysis using median time lags of 4 and 3 weeks was also run using lower bound VE (50% and 70% for first dose and complete series, respectively) and upper bound VE values (70% and 97.5%, respectively). We deemed this range of values representative of the observational studies for the vaccines most frequently used in the countries in this study (  The number of observed deaths plotted here are the true numbers reported by the country and not the rolling average used in calculations.

Table 2
Age group and country breakdown of full vaccination uptake, number of deaths observed and averted as well as observed and expected mortality rates per 100,000 population, in the population aged 60 years and older, using the base vaccine effectiveness scenario a , by country, 30 WHO European Region countries with sufficient data, weeks 51/2020-45/2021 129,851 and 733,744, respectively, in people 60 years and older in the 33 countries (Table 1, Supplementary  Table S1). The proportion of total estimated deaths averted by vaccination ranged between 23% and 62% according to the VE sensitivity analysis (Supplementary  Table S2, where also country-specific estimates are available). There was minimal variation in deaths averted by varying the time lag (Table 1, Supplementary  Table S1) .

Discussion
Overall  [3,4,17,18], or estimated the difference in life expectancy in three countries [19]. All showed the positive impact of vaccination in saving lives. Our analysis adds to these studies by using a standard approach to compare the estimated direct impact of differential roll-out of COVID-19 vaccine programmes across 33 diverse countries in older adults across the European region.
The analysis contains several limitations and assumptions. Our estimate is conservative as we did not estimate the indirect effect of vaccination or the impact of public health and social measures on mortality by reduction in transmission. The capacity of healthcare systems to respond to the pandemic were not considered here. These primary findings are derived using use a base scenario of vaccine effectiveness and time lag. Using more extreme vaccine effectiveness estimates against mortality in particular scenarios can vary the number of deaths averted as shown in the sensitivity analysis. VE was not differentiated by vaccine manufacturer or type. Based on current, limited data, we assumed that VE against mortality was similar for the SARS-CoV-2 Alpha and Delta variants (Phylogenetic Assignment of Named Global Outbreak (Pango) lineage designation B.1.1.7 and B.1.617.2), the predominant circulating variants of concern during our study period. In addition, we assumed that there has not been any waning in protection against severe disease to date [2].
Differential reporting in mortality surveillance systems was another potential limiting factor, which we assumed to be comparable between countries. Additional third doses of vaccination have not been considered here, as countries had only started to implement them. Many of these limitations and assumptions are likely to have under-estimated the true number of deaths in each country to varying extents. We attempted to account for this by undertaking an 'observed over expected' analysis, estimating the proportion of deaths averted by vaccination to enable a more robust comparison of direct vaccine impact between countries.

Conclusion
We show that since the start of COVID-19 vaccination in Europe, the lives of many older adults have been saved through immunisation. Our results highlight large differences between countries in the direct impact of vaccination on COVID-19-related mortality. Early and complete implementation of vaccination of older adults was associated with the largest reduction in expected deaths. It is critically important that all countries rapidly achieve high coverage for these priority groups to prevent further morbidity and mortality, particularly with SARS-CoV-2 transmission rates increasing as Europe moves into the 2021/22 winter period.