Understanding the evolution of a pandemic: what is revealed by the data from the first two waves of COVID-19 in Italy
A study conducted by the Sant'Anna School of Advanced Studies in Pisa through international collaborations examined the two pre-vaccine waves of COVID-19 in Italy (between February 2020 and February 2021). The results showed how analysing territorial data with innovative techniques could support traditional epidemiological monitoring
A scientific study on the first two waves of COVID-19 in Italy—the pre-vaccine waves between February 2020 and February 2021—proposes an innovative approach to monitoring and interpreting the evolution of the pandemic at the territorial level, offering useful tools to understand the dynamics of epidemic phenomena in the absence of vaccine-induced immunity. The study, published in Scientific Reports, was conducted by the Scuola Superiore Sant’Anna in Pisa in collaboration with several research centers in Europe and North America (IBM Research Europe, Emory University, Carnegie Mellon University, Université Laval, and Penn State University).
By employing statistical techniques not traditionally used in epidemiology, the study results confirm the importance of timely restrictions in reducing the mortality rate associated with infections, as well as a link between local mobility and mortality in both pre-vaccine waves.
“The social and political debate on the costs and benefits associated with restrictions and mobility controls is of fundamental importance, and in the years following the pandemic,
Data from 107 Italian provinces and functional data analysis
The study analysed the first two waves of COVID-19 in Italy, using provincial-level data on mortality, mobility, timing and intensity of government restrictions, as well as socio-demographic, infrastructural and environmental factors. The study used advanced functional data analysis techniques, which allow robust trends to be identified even in the presence of incomplete data.
“Functional Data Analysis techniques allow us to compare phenomena that occur asynchronously or in different time frames, and to incorporate dynamic and static variables. Moreover, they do not rely on the assumptions used in standard epidemiological models, and thus bring a more flexible and data-driven perspective to the analysis” explains Chiaromonte.
The two pre-vaccine waves differed significantly in terms of mortality patterns. The first had higher, more concentrated mortality peaks, while the second was more widespread and asynchronous across the country. However, in both waves Italian provinces formed three distinct clusters; one characterized by mild mortality and two characterized by intermediate and harsh exponential patterns.
“Interestingly, locations most impacted during the first pre-vaccine wave were among the least impacted during the second – says Chiaromonte – Understanding the causes of this pattern would require disaggregated data that are not available to us, but potential factors may include a reduction in the number of vulnerable individuals following deaths in the first wave, behavioural adaptations leading the population to adhere more strictly to recommendations (e.g., social distancing and mask use), and perhaps to a degree of herd immunity”.
Restrictions and mobility: a key role in containing mortality before vaccines
The study highlights how timeliness in introducing restrictions was one of the factors most closely associated with mortality during the first two waves of the pandemic, prior to the arrival of vaccines. The analysis suggests that, statistically, restrictive measures contributed significantly to reducing the impact of the epidemic. The two waves showed notable differences in mobility behaviours: during the first lockdown, a marked and uniform reduction in movements was observed across the entire country, whereas during the second wave—characterized by the tiered, color-coded system—reductions were more limited and varied across different areas. Despite this, in both phases a clear statistical association emerges between higher levels of mobility and increased mortality.
“The work that led to this article is a wonderful example of collaboration among Italian scientists who cultivate and use their competences across the globe” adds Chiaromonte.
Study References
Tobia Boschi, Jacopo Di Iorio, Lorenzo Testa, Marzia A. Cremona & Francesca Chiaromonte, Contrasting pre-vaccine COVID-19 waves in Italy through functional data analysis, Scientific Reports, 2025
Study link: https://www.nature.com/articles/s41598-025-29316-4
The study was supported by: the L'EMbeDS Department of Excellence of the Sant'Anna School of Advanced Studies, the SMaRT COnSTRUCT project (part of FAIR, Italian PNRR funded by NextGenerationEU), the Huck Institutes of the Life Sciences at Penn State, the Natural Sciences and Engineering Research Council of Canada (NSERC), the Fonds de recherche du Québec Santé (FRQS), the FRQ-IVADO Chair in Statistical Learning, and the Faculty of Business Administration at Université Laval.
