Modeling the spread of COVID-19 in New York City and the impact of socio-economic inequalities

Since the start of the COVID-19 pandemic, there has been an increasing awareness about the geographical heterogeneity of the spread of SARS-CoV-2 and such heterogeneity has been observed at different spatial scales in many countries across the world. This thesis investigates the possible relations between the geographic disparities in COVID-19 incidence and the differences in socio-demographic and economic profiles of the New York City neighborhoods. Moreover, the objective of the work is also to study the effects of the interventions imposed by the NYC government on the infection spread, in the period from March to October 2020. To this aim, we use a computational approach to model the epidemic dynamics in NYC at the zip code level. A network structured meta-population model is implemented to reproduce the geographical transmission trends observed in NYC. The advantage of granular spatial modelling of the meta-population model is combined with the possibility to realistically describe the variations of mobility flows across NYC neighborhoods in the early phase of the pandemic. Mobility patterns are included in the model thanks to available mobile phone data at the census tracts level. The effects of the differential behavioral response to social-distancing policies are summarized in the model by a local risk factor, which affects the transmissibility of the infection at the spatial level in the simulations. Our results show that the simulated geographical patterns of the COVID-19 incidence correctly reproduce the actual observed spatial trend of the epidemic across NYC neighborhoods, confirming the fundamental role of human mobility and socio-economic factors in the epidemic.