Background: The number of confirmed COVID-19 deaths differed across countries and across waves of the pandemic. Patterns also differed between groups within a country.

Objective: We combine data on excess mortality with data on cause-of-death-specific mortality in the case of Sweden to identify which groups had excess mortality beyond what can be captured by analyses of COVID-19-specific deaths. We also explore the possibility that some groups may have benefited in terms of reduced all-cause mortality, potentially due to home-centered living conditions during the pandemic.

Methods: We produced and compared three sets of group-specific incidence rates: deaths from (1) any cause in 2020, (2) any cause in 2019, (3) any cause excluding COVID-19 in 2020. We compared rates across different socioeconomic profiles based on combinations of sex, age, marital status, education, and country of birth.

Contribution: We show that many of those who died during 2020 would not have done so in the absence of the pandemic. We find some evidence of COVID-19 mortality underestimation, mainly among individuals with a migration background. We also found groups for which mortality decreased during the pandemic, even when including COVID-19 mortality. Progression across the first and second waves of the pandemic shows that more groups appeared to become protected over time and that there was less underestimation of COVID-19 mortality in the second part of 2020.