Understanding the uneven spread of COVID-19 in the context of the global interconnected economy

The COVID-19 coronavirus disease (SARS-CoV-2) is the new pandemic that emerged in December 2019, in Wuhan city, China, and ever since has been rapidly spread around the world, causing cascading deaths to humanity, vast pressures on the national public health systems, and uncertainty about the future of the global and national economies1,2. The peculiar biological, epidemiologic, and spreading COVID-19 features3,4,5,6,7 have equipped this disease with the threat properties of a pandemic, resulting to the global awareness and outbreak of research. It has been almost two years since its emergence in Wuhan, and the COVID-19 pandemic has already become a prime concern and priority for the scientific community2,5. According to the Google Scholar academic database8, the search of the keyword “COVID-19” yields approximately 4.35 m (million) results, whereas other keywords referring to established research fields (many of which enjoy centuries of scientific research) yield a comparable number of results, such as the cases of “gravity” (4.06 m results), “cancer” (6.05 m results), “networks” (5.64 m results), “economy” (4.83 m results), “electric” (6.69 m results), “health” (6.93 m results), “space” (7.3 m results), and “science” (8.44 m results). In the already vast COVID-19 literature, someone can observe three major research strands (directions) in the study of the pandemic’s spread. The first concerns the patterns and causes9,10,11,12,13, the second strand regards the effects2,14,15,16, and the third one the management, treatment, and cure17,18,19,20 of the COVID-19 spread. These directions involve applications in various fields, such as individual and public health, medicine and clinical research, pharmacy, policy, economy, society, education, communication, transportation, environment, geography, and many others. According to all these diverse approaches, relevant scientific research is multidisciplinary, broad, and complex2,5,7,21,22,23, illustrating the linkage of COVID-19 with all aspects of everyday life, along with the intention of the humanity to get ahead in the fight against the pandemic.

Although a thorough review of the vast and multidisciplinary COVID-19 literature is an ongoing and future challenge for epistemology researchers, the relationship between the pandemic’s spreading and the interconnected socioeconomic structure of the modern world is more than evident in the literature2,12,16,24,25,26,27. At the microscopic level, the relationship between individual connectivity and COVID-19 spread (transmission) builds on clinical and epidemiologic terms. This approach has already enjoyed fruitful research contributing to the understanding and pandemic management28,29,30. In macroscopic terms, the effect of interconnectedness on the pandemic’s spread is mainly studied on a dual basis, either within or between countries (in a cross-country framework). The first approach covers all topics of interest about the pandemic, such as regional outbreaks and spread due to imported cases31,32, mobility, and travel restrictions14,15, the effects of lockdown17,33, and others4,34. Even when is implemented on an international scale, the within-countries approach conceives interconnectedness as an intrinsic property of countries, interpreting the uneven spread of the pandemic based on comparing such intrinsic properties between countries. In addition, the second (cross-country) approach conceptualizes the spread-channels (links) of the pandemic between countries by configuring variables or indicators approximating aspects of interconnectedness, such as the number of tourists, geographical distance, exports per capita, motorway density, etc.35,36,37.

However, interconnectedness becomes better conceivable in the context of communication theory38,39, according to which the infection of COVID-19 is the result of virus transmission between two individuals3,40, namely an elementary communication structure called either link or edge41. Building on such pair-wise configuration, the complex COVID-19 virus transmission system can excellently represent a network structure and be modeled as a graph, as conceived by network science41,42,43,44, at any spatial and functional level. In the literature, this cognition appears to become deeper. For instance, the authors of45 applied social network analysis to track the spread of COVID-19 in India and found that international travels were determinative of the pandemic outbreak in the country. On the same basis, the work of46 studied a social network of the COVID‑19 spread in South Korea, finding that the topology of the infection network was dependent on the policy measures applied by the government to control the pandemic. The paper of47 constructed a multilayer network of social connectivity to study the COVID-19 epidemic spread in Brazil and the country’s potential to manage the effect of the pandemic, proposing the increase of social isolation (up to a lockdown) as the best option to preserve the functionality of the healthcare system capacity. The work of48 developed a multilayer network methodology of the relationship between socio-cultural and economic characteristics to identify the pandemic spreaders, resulting in a classification of the countries according to their socioeconomic, population, Gross Domestic Product (GDP), health, and air connectivity attributes. The authors of49 developed a network model of social contacts between individuals, calibrating their model on data about COVID-19 spread in Wuhan (China), Toronto (Canada), and the Italian Republic, using a Markov Chain Monte Carlo (MCMC) optimization algorithm. The results showed a good fitting of the network model to empirical data and provided a tool for the health authorities to plan control measures against the pandemic. The paper of13 used social network analysis to identify the clusters of the COVID-19 global spread and observed that the pandemic spread followed a route from China to West Asia, Europe, North America, and South America according to three classifications. The authors of50 developed a SAIR (Susceptible-Asymptomatic-Infected-Removed) model on social networks to study the spread of COVID-19 and examined the effectiveness of policy measures aiming to control the pandemic. Also, the authors of51 used agent-based simulations based on the SEIR (susceptible–exposed–infectious–recovered) model to show that two network intervention strategies dividing or balancing social groups can substantially reduce transmission while sustaining economic activities. On a similar framework, the authors of52 developed a contact network to study the city-level transmission using an infectious agent spread (SEIR) model, showing that precise knowledge of epidemic transmission parameters is not required to build an informative model of the spread of disease.

Further, the work of53 studied the impact of human mobility networks on the COVID-19 onset in 203 different countries. The authors used exponential random graph models to perform an analysis of the country-to-country global spread of COVID-19. The analysis showed that migration and tourism inflows were factors increasing the probability of COVID-19 case importations. The authors of54 studied a knowledge network model of semi-supervised statistical learning constructed on aerial mobility data of Hong Kong and Wuhan. The purpose of the study was to determine the early identification of infectious disease spread via air travel and align the need to keep the economy working with open connections and the different dynamics of national pandemic curves. The work of55 applied a network inference approach with sliding time windows to capture the spatiotemporal influence of infections and trace the spread of the pandemic in New York and the USA. The paper of56 employed an agent-based model to nearly 1.6 million firms in Japan and simulated the pandemic’s propagation, where they evaluated lockdown scenarios of Tokyo, Japan. On a similar conceptualization, the authors of57 examined how the economic effects of lockdowns in different regions interact through the supply network of 1.6 million firms in Japan. The analysis showed that a region’s upstream-ness, the intensity of loops, and supplier substitutability in supply chains with other regions largely determine the economic effect of the region’s lockdown. The paper of58 investigated the impact of COVID-19 on global air transportation at different geographical scales, namely worldwide, international country, and domestic airport networks, for representative cases, and found that the impacts of the pandemic were concordant, in intensity, with the geographical scale. The work of59 developed a complex network of COVID-19 correlations between 122 countries and empirically investigated the network connectedness influencing macroeconomic and social factors. The analysis showed that population density, economic size, trade, government spending, and quality of medical treatment are significant macroeconomic factors affecting COVID-19 connectedness in different countries.

As is evident from the previous review, the network paradigm provides an insightful approach for studying and understanding the spreading of the pandemic within the context of the modern world’s connectivity. This approach goes beyond other non-network counterparts, which conceive interconnectedness either as an intrinsic property of countries4,27,33 or in terms of variables or indicators approximating aspects of cross-country interaction35,36,37. Therefore, using the network paradigm allows developing graph models representing interconnected structures at various geographical scales. However, there is still a long way to go on the network analysis of the interconnectedness of COVID-19 because current empirical approaches are relatively restricted compared to the broad range of human economic activity. In particular, the relevant research mainly applies to (a) social network models45,46,47,48,49,50,51, (b) supply and logistic networks56,57, and (c) networks of human mobility53,54,60. Moreover, due to big-data demand, relevant studies implemented at the global scale are considerably fewer. Toward responding to this demand, this paper focuses on the patterns and causes (first strand) of the COVID-19 spread. It develops a multidimensional methodological framework for understanding the spatio-temporal spread of the pandemic in the context of the global economy modeled as an interconnected cross-country structure. Therefore, this study goes beyond the previous complex network approaches by constructing a global network model incorporating dimensions of topology, geography, and economic openness. To do so, it conceptualizes worldwide interconnectedness based on economic globalization61 and, specifically, by constructing a network model of international tourism flows.

In terms of transport geography62 and spatial economics63, a tourism network belongs to the family of transportation networks. However, it is quite different from a typical transportation network, such as road, railway, maritime, or air transport network43,64,65,66, both in terms of structure and functionality. First, a tourism network has a specific economic configuration of its transport demand64,67,68, which involves movements for tourism purposes outside the country of residence. On the contrary, in a typical transportation network, the economic forces driving the transport demand are broad and unspecified since they may refer to trade, occupation, recreation, health, education, etc.62,69. Therefore, as far as the economic configuration of transport demand is concerned, a tourism network is well-defined within a unified functional framework (or economic activity), while a typical transport network (except a multilayer model) is not and is multivariable. However, in terms of modal configuration, a tourism network is multimodal because movements may occur through all transportation modes62. In contrast, a typical transport network usually is defined within a single transportation mode43. Finally, in terms of network topology43, a tourism network is more topological than geometric64,67,68, as a typical transportation network usually is. In particular, in a tourism network, edges are usually defined as accessibility links and not as routes or transportation channels62,67. Within this context, the network model of international tourism flows constructed in this paper suggests a good proxy35,45,53 for the outbreak of the pandemic because (a) it provides a comprehensive economic setting (tourism), (b) is integrated in terms of modal configuration, and (c) is more representative in terms of network accessibility.

This study builds on a three-dimensional conceptual model to analyze the worldwide spatio-temporal spread of COVID-19. It incorporates one dimension approximating the interconnectedness of the international tourist mobility, a second one describing the openness of countries to the globalized economy, and a third one expressing the spatial impedance to transportation. By constructing a single network model, this paper proposes an integrated framework for the study of the spatio-temporal spread of COVID-19. It also contributes to the literature with more realistic models of the worldwide interconnected system, where COVID-19 and other pandemics are spreading. The remainder of this paper is structured as follows: second section presents the methodological and conceptual framework of the study, third section shows the results of the analysis and discusses them within the context of regional and geographical sciences, and finally, in last section, conclusions are given.

Leave a reply:

Your email address will not be published.