Multi-scale approach to disease spreading in social net- works Social scientists have long identified a small number of underlying social mechanisms which individuals follow in the course of interactions with each other. In this project we propose to perform a multi-scale analysis to investigate how these micro-level social mechanisms of individual actions are transformed into network patterns at the large-scale that, in turn, determines the spread of infectious diseases in a social network. We will develop a systematic study of individual interactions in social communities, with the goal of understanding the process of establishing new network ties. This will reveal the global structure of social network as the support for the spreading of infectious diseases. This information will be later used to lower the threshold of immunization by identifying those individuals who, through their position in the network, are more likely than average to spread the disease to the larger portion of the social network. The novelty of our project lies in the multi-scale approach, where we continuously transfer information from local social mechanisms to global network properties, according to the following layers: individual actions (micro- scopic level)! global network of connections (macro-level)! disease spreading dynamics to identify the fastest spreaders of disease (superspreaders). Our approach has the novel value of studying the real-time dynamics of connections where we can directly monitor exact timing of every new tie in the network, i.e., how this dynamics affects the spread of disease. The network of connections in real communities will be analyzed in terms of state-of-the-art network theory. A large number of valuable tools from statistical network analysis will inform us about the multi-scale structural properties of the social networks and how these properties affect the transmission of disease in a network. The innovation of the proposal lies in the introduction of novel Systems Science/Statistical Physics tools to sociologically- relevant health problems. The project is a collaboration between a physicist with expertise in Complex Science (Makse) and a sociologist (Liljeros) who will provide the wide range of expertise in multi-disciplinary domains necessary to evaluate the results from scientifically varying standpoints.
We will perform a multi-scale social network analysis developing state-of-the-art mathematical tools from Systems Science and Statistical Physics. The impact on public health is significant, given that we will try to uncover and model how social connections are formed and how we can use this information in order to identify the individuals, named superspreaders, who are most likely to contribute to the largest spreading of disease in a social network. Our results will have direct applications in behavioral research and will contribute to a better understanding of how disease spread in contact social networks with applications to the design of efficient immunization strategies. Identifying superspreaders according to their position in the network structure will allow to decrease the number of people who must be vaccinated to immunize a community against an infectious disease. This information could save resources and decrease outbreak sizes.