Covid-19 related excess mortality: An analysis by age for selected countries

BACKGROUND: Measures of excess mortality are generally regarded as providing the best metric for countries' experience during the Covid pandemic of 2020-2022. Such measures, however, need to be adjusted to take account of the different age structures of different countries as Covid-related mortality can differ substantially by age. METHODS: This paper analyses excess mortality data for a selection of countries using the Short Term Mortality Fluctuations dataset. A standardised age distribution is applied to the data to obtain age-adjusted excess mortality and this is compared to crude excess mortality data. The metric of years of life lost (YLL) is also calculated for the same dataset and cluster analysis is applied to investigate relationships between the mortality metrics and GDP per capita and inequality. RESULTS: The age adjustment makes a significant difference for some countries but the overall ranking of countries in terms of excess mortality changes very little, with rank correlation coefficients in excess of 0.95 for the metrics of crude mortality, age adjusted mortality and YLL Age adjusted excess mortality has a rank correlation of -0.69 with GDP per capita but the rank correlation with the Gini coefficient of 0.11 is not statistically significant. CONCLUSIONS: Rankings of countries by excess mortality during the Covid period are robust to age adjustment and choice of mortality metric. Cluster analysis suggests distinct clusterings of countries with similar outcomes, particularly the former planned economies of central and eastern Europe.