Almost all studies of the health impact of COVID-19 have focused on analyzing incomplete and often inaccurate data on reported infections, hospitalizations and deaths. We are estimating excess deaths by age and sex and over time. We study excess deaths caused directly by COVID-19 or indirectly because of pressures on the health care system and effects of policy interventions. These excess deaths are a measure of the total impact of the COVID-19 pandemic. We also estimate the life-years-lost because of the excess deaths. To gain a deeper understanding of the nature of the COVID-19 pandemic, we analyze excess deaths by municipality, by living arrangements, and by socio-economic status. We compare Denmark to Sweden, and later to Norway, Iceland, Italy and Spain with regard to the impact on policy interventions on excess deaths and life-years lost. The project has two research strands:

•  The first focuses on Denmark: beyond the estimates of excess death and life-years lost for the whole country, the project will look at the spread and control of COVID-19 in Danish municipalities. It will also focus on spread within households, including collective living in nursing homes. Use of individual data will permit analysis of excess mortality and life-years lost by socio-economic status, which is a major concern because COVID-19 may widen social disparities in length of life.
  
  
• The second research strand compares Denmark with Sweden, Norway, Italy and Spain. This will permit analysis of the effectiveness of alternative policy responses on the two key measures analyzed by the project —number of deaths and number of life-years lost, directly or indirectly due to COVID-19. 

“We develop radically different strategies to estimate excess deaths directly and indirectly due to COVID-19. The problem is so important and complicated that more than one method for estimating excess deaths should be developed and tested,” stresses Professor James Vaupel, the Principal Investigator. He further specifies his plans: “The basic method used by infectious disease epidemiologists makes strong assumptions about functional forms—using, for instance, sine curves and quantities raised to 2/3 powers—and requires estimation of many parameters. We will pursue more flexible, less restrictive methods based on our work on penalized splines, on Seasonally adjusted AutoRegressive Integrated Moving Average (SARIMA) modeling, on models of mortality in heterogeneous populations, and on methods for forecasting age-specific death rates.”

Note: The project was conceived and led by James W. Vaupel. After his sudden death, the role of principal investigator was taken over by Silvia Rizzi.

The project Excess deaths and life-years lost due to the Covid-19 pandemic in Denmark is supported by the ROCKWOOL Foundation.