Research into COVID-19 excess mortality helps us prepare better for future pandemics
During the crisis, it became clear how profoundly pandemics can affect our demographic models. These models form the foundation for policies in public health, pensions, and insurance. How can we improve them so that we are better prepared for the future?
Researchers Frank van Berkum, Bertrand Melenberg, and Michel Vellekoop set out to answer this question. In this collaboration between the Research Centre for Longevity Risk (part of the University of Amsterdam) and Tilburg University, they developed a new framework to better quantify excess mortality during pandemics and to forecast mortality under various scenarios for the future development of such pandemics.
The study is based on earlier work that is incorporated in the most recent mortality prognosis of the Dutch Actuarial Association. That prognosis is used by most Dutch pension funds and insurance companies, and it helps these institutions to ensure that they have enough reserves to protect pensioners and policyholders against the financial consequences of economic and demographic shocks.
Better predictions and better scenarios
By combining historical mortality figures with detailed weekly data from the COVID-19 period, the researchers were able to pinpoint exactly how the pandemic affected different age groups. This enables policymakers and public health authorities to make much more accurate estimates of excess mortality and its age distribution — and to better evaluate whether measures and interventions have been sufficiently effective.
The study also contributes to both short- and long-term perspectives for actuaries and demographers, allowing them to make more precise assessments of how a pandemic impacts life expectancy. This in turn leads to more accurate scenarios for the financial consequences of pandemics, providing policymakers with a stronger empirical basis for their decisions.
By using detailed data and robust statistical models, we can paint a clearer picture of the real impact of pandemics on mortality — and support better decisions for society as a whole.Frank van Berkum, researcher Research Centre for Longevity Risk
Combining large datasets
The researchers used existing forecasting models as a starting point but adapted them to the unique circumstances of a pandemic. They first mapped general, long-term trends by analysing decades of mortality data. They then integrated this with the much more detailed weekly figures collected during the COVID-19 pandemic by Statistics Netherlands (CBS). This approach allowed them to closely monitor changes and to reveal clear differences between age groups. By combining long-term trends with high-frequency, real-time data, the team developed a statistical model that accurately reflects the pandemic’s effects — and that can also be used to explore future scenarios.
Developing adaptive models
The researchers recommend that future studies not only include data from the pandemic itself, but also from the post-pandemic period — the phase we are currently living through. One open question, for instance, is whether mortality rates will return to pre-pandemic trends, or whether the pandemic has had a lasting impact. They also express the ambition to develop adaptive models that can flexibly respond to new data from the Netherlands and other countries. Future research could additionally focus more on specific causes of death, to provide an even clearer and more complete picture of demographic developments.
By connecting knowledge and data, these researchers not only help us understand the past - they also ensure that we are increasingly well prepared for future pandemics.