Severe features during outbreak but low mortality observed immediately before and after a March–May 2020 COVID-19 outbreak in Stockholm, Sweden

Ville N Pimenoff,Mikael Björnstedt,Joakim Dillner

doi:10.1016/j.ijid.2021.08.005

Abstract

ObjectivesThe aim of this study was to estimate how well the excess mortality reflected the burden of coronavirus disease 2019 (COVID-19)-related deaths during the March–May 2020 COVID-19 outbreak in Stockholm, Sweden, and whether the excess mortality during the outbreak might have resulted in a compensatory reduced mortality after the outbreak. MethodsUsing previous 10-year or 5-year average mortality rates as a baseline, the excess mortality estimates before, during, and after the COVID-19 outbreak in March–May 2020 in Stockholm were compared. ResultsWeekly death estimates revealed that the immediate pre-outbreak and post-outbreak all-cause mortality did not exceed to excess mortality regardless of whether previous 10-year or 5-year average mortality was used. Forty-three days after the start of the outbreak, 74.4% of the total excess mortality was reportedly explained by known COVID-19-related deaths, and the present study reports an update, showing that 15 weeks after the start of the outbreak, the reported COVID-19-related deaths explained >99% of the total excess mortality. ConclusionsAn exceptional outbreak feature of rapid excess mortality was observed. However, no excess but similarly low mortality was observed immediately prior to the outbreak and post-outbreak, thus emphasizing the severity of the first wave of the COVID-19 outbreak in Stockholm.

Highlights

Excess all-cause mortality is increasingly being used to monitor the timing and severity of COVID outbreaks; the COVIDattributable fraction of the all-cause mortality is not always clear
We and others have previously reported that the excess mortality estimates tend to exceed the COVID-19-attributable mortality data (Felix-Cardoso et al, 2020; Woolf et al, 2020), with only 74.4% of the excess mortality being attributable to COVID-related deaths in Stockholm when including the reports that had arrived up to 43 days after the start of the local outbreak (Pimenoff et al, 2020)
2020), excess all-cause mortality can give a more timely picture of the burden of outbreaks associated with a substantial increase in mortality rates, such as COVID-19, compared with early reporting of COVID-19-attributable deaths

Summary

Introduction

Excess all-cause mortality is increasingly being used to monitor the timing and severity of COVID outbreaks; the COVIDattributable fraction of the all-cause mortality is not always clear (www.euromomo.eu). When calculating weekly excess mortality, fluctuations in mortality rates by calendar year are minimized by calculating a baseline mortality over several years (Ledberg, 2020). As there are temporal trends in mortality (increased life expectancy), the average mortality estimation cannot use too many years (Ledberg, 2020). The aim of this study was first to assess how well the excess mortality reflected the burden of COVID-19-related deaths during the March–May 2020 COVID-19 outbreak in Stockholm, Sweden, and to investigate whether the excess mor-. Tality during the outbreak might have resulted in a compensatory reduced mortality after the outbreak

Objectives

Methods

Conclusion