While the COVID-19 pandemic disrupted societies worldwide, its demographic toll in France was far lower than the 1918 Spanish flu epidemic or the 1911 heatwave. Florian Bonnet and Carlo-Giovanni Camarda compare five major longevity crises over the past century to put COVID-19’s burden into historical and spatial perspective.
How severe was the demographic impact of COVID-19 in the first two years of the pandemic? Many articles published in recent years have sought to answer this question by comparing different national experiences. We ourselves contributed to this effort by publishing two articles examining excess mortality across a large number of European regions in 2020, and then in 2020 and 2021 (Bonnet at al. 2024a,b). However, understanding what makes COVID-19 unique requires not only international comparisons, but also a longer-term historical perspective to compare its burden with those of other longevity crises.
In a recent article, we assessed the historical and spatial burden of COVID-19 in France by comparing it to four major longevity crises since 1900: the Spanish flu epidemic (1918–1919), the Hong Kong flu epidemic (1968–1969), and the heatwaves of 1911 and 2003 (Bonnet and Camarda 2025). Our study uses annual death and population counts for 90 French departments drawn from the French Human Mortality Database, combined with statistical modeling techniques, to estimate excess mortality related to each crisis.
ASYLL, or age-standardized years of life lost
To compare these events across space and time, the choice of metric is crucial; we used age-standardized years of life lost (ASYLL) as the central indicator of our study. Widely used in public health, this metric quantifies the total years of life lost due to premature death, and can be applied to estimate the burden of specific crises. One of its main advantages is that it enables meaningful comparisons between different populations by relying on a common age structure. Another key benefit is that it can be summed across multiple years, allowing researchers to capture the full impact of a pandemic that spans more than a single year – something that is not possible with other measures, like drops in life expectancy. Finally, it provides a more nuanced picture of the mortality burden than the number of deaths, by considering the age at which deaths occur.
Ranking mortality crises
Our findings show that COVID-19 led to an average of 9 years of life lost per 1,000 inhabitants in France over 2020–2021. How high is this in historical terms? Measuring in ASYLL per 1,000 population,
• the 2003 heatwave caused about 2.5 years of life lost,
• the 1968–69 Hong Kong flu between 5 and 7,
• COVID-19 about 9 (as mentioned above),
• the 1911 heatwave close to 24, and
• the Spanish flu nearly 100.
These figures help contextualize the scale of COVID-19’s mortality impact. Despite its global scope and the societal disruption it caused, the demographic burden of COVID-19 in France remains about 10 times lower than that of the most devastating crisis of the 20th century, the Spanish flu epidemic (excluding the two world wars, to be sure).
The age profile of deaths varied markedly between crises. Excess mortality related to COVID-19 and the 2003 heatwave was heavily concentrated among adults aged 60 and over. In contrast, the Spanish flu disproportionately affected children and young adults. Finally, the heavy toll of the 1911 heatwave was largely driven by infant deaths, in many cases stemming from contaminated milk. These age patterns partly explain why the overall burden of the COVID-19 pandemic, as measured by ASYLL, appears lower than that of other longevity crises, especially those of the first half of the 20th century: premature deaths at younger ages result in more years of life lost due to the higher remaining life expectancy at those ages, an outcome that would also emerge with other metrics, though ASYLL captures it more directly.
Geographical distribution
As Figure 1 shows, in each of the five cases considered, striking regional contrasts emerged, with some crises showing widespread effects across the country and others confined to specific areas.
• The Spanish flu was remarkably uniform in its spread, affecting all French regions with comparable intensity, although slightly less in the north-west of the country.
• The 1911 heatwave, in contrast, affected northern and south-eastern France especially, with particularly severe losses in the Lozère department (84 years of life lost per 1,000 population).
• The 2003 heatwave had the strongest impact in the central-western part of the country, including departments such as Creuse, Indre and Loir-et-Cher (about 10 years of life lost per 1,000 population).
• COVID-19 stands out for its highly fragmented geography. Mortality was particularly severe in the Paris region (Île-de-France) and eastern regions bordering Belgium, Germany, and Switzerland (about 20 years of life lost per 1,000 population). Parts of western France experienced little to no excess mortality, especially in 2020 (Bonnet et al. 2024b).
This geographical heterogeneity is not simply a product of random variation; it underscores the crucial role of regional and local dynamics in shaping health outcomes, particularly during a global pandemic. National-level aggregates often conceal significant internal disparities, which public authorities must recognize in order to better allocate resources and design effective interventions. In our article (Bonnet and Camarda 2025), we found that the 2003 heatwave showed the highest territorial diversity, followed by COVID-19; the Spanish flu, on the other hand, had the most geographically uniform impact on mortality. The marked variability in the impact of the 2003 heatwave is unsurprising, given that mortality rates were highest in areas where peak temperatures were most extreme, especially in the central-west region. In contrast, the notable regional disparities observed during the COVID-19 pandemic are likely attributable to stringent social distancing measures implemented during the first wave (Spring 2020), which limited the virus’s spread across the national territory, protecting regions in the west and south especially. In comparison, the minimal social distancing measures during the Spanish flu epidemic were compounded by the depleted state of the health system in the aftermath of the First World War that exacerbated the pandemic’s impact. Additionally, the population’s familiarity with death in the post-war era further hindered effective containment efforts, allowing the virus to spread uniformly across the French regions.
Conclusion
Placing the COVID-19 pandemic within a century-long perspective shows that its impact – while unprecedented in terms of media coverage and social disruption – is part of a longer history of unevenly distributed longevity crises.
References
Bonnet, F., Camarda, C.G. (2025). Contextualizing the Global Burden of COVID‐19 Pandemic: A Historical and Geographical Exploration of Excess Mortality in France, 1901–2021. Population and Development Review.
Bonnet, F., Grigoriev, P., Sauerberg, M., Alliger, I., Mühlichen, M., Camarda, C.G. (2024a). Spatial variation in excess mortality across Europe: a cross-sectional study of 561 regions in 21 countries. Journal of Epidemiology and Global Health, 14(2), 470-479.
Bonnet, F., Grigoriev, P., Sauerberg, M., Alliger, I., Mühlichen, M., Camarda, C.G. (2024b). Spatial disparities in the mortality burden of the covid-19 pandemic across 569 European regions (2020-2021). Nature Communications, 15(1), 4246.