Led by Dr. Alexander Kirpich, Dr. Pavel Skums, and Dr. Alexander Perez Tchernov, the research collaboration developed and applied a clustering-based methodology to examine whether countries with similar COVID-19 dynamics also share comparable public health and sociodemographic characteristics. The study analyzed data from 42 European countries and focused on six variables: COVID-19 incidence, mortality, vaccination rates, SARS-CoV-2 genetic diversity, cross-country mobility, and selected sociodemographic indicators.
The methodological framework involved calculating pairwise distances between countries for each variable and constructing hierarchical clustering trees. Relationships between the resulting cluster structures were quantified using cophenetic correlation and Baker’s Gamma correlation, allowing for systematic comparison across variables.
The analysis demonstrated how cluster-based methods can be used to assess temporal epidemiological data in relation to broader structural and demographic factors. The findings showed that vaccination clustering had moderate agreement with incidence, but limited correspondence with mortality. Mortality-based clustering was aligned only with population health indicators. Incidence-based clustering corresponded with patterns in genetic diversity, population health, and sociodemographic characteristics. Genetic diversity clustering also showed notable alignment with cross-country mobility and related demographic variables.
This work emphasizes the utility of clustering techniques in public health research, particularly for identifying structural similarities across countries based on time-dependent epidemiological and contextual data.
The findings were presented at the ITSE 2025 Conference in Gran Canaria, Spain, in July 2025, and the full manuscript was published in Royal Society Open Science in September 2025.