The novel coronavirus SARS-CoV-2 and the disease it causes, COVID-19, triggered a global pandemic. Although understanding international transmission dynamics is essential, inferring diffusion networks from observational data remains challenging due to their inherent complexity. In this study, we investigated the global transmission process, including temporal delays, using worldwide COVID-19 case data from January 3, 2020, to December 31, 2022. We analyzed the data using Complex Hilbert Principal Component Analysis, which captures both concurrent relationships and lead-lag dynamics. We then examined interactions among countries with respect to six factors: geography, population, GDP, stringency of countermeasures, vaccination rates, and government type. The results reveal that two primary trends coexisted throughout the period, one in 2020 and another in 2021–2022, with their dominance alternating over time. Specifically, in 2020, European, high-income, and democratic countries led the first trend and were typically associated with higher transmission levels. In contrast, in 2021 and 2022, countries in Africa and the Americas, particularly those with lower income levels, emerged as leading contributors to the second trend. We also found that, while internal countermeasures may have helped suppress domestic infection levels, they did not affect the international transmission pattern. Furthermore, although vaccination became widespread in 2021, it did not alter the pattern of international spread.
Hiroyasu Inoue (Mon,) studied this question.