Researchers at Columbia University’s Mailman School of Public Health used computer simulations to reconstruct how the 2009 H1N1 influenza pandemic and the 2020 COVID-19 pandemic expanded across U.S. metropolitan areas. The study, published in the Proceedings of the National Academy of Sciences, found that both viruses were already circulating widely in many cities within weeks, with air travel playing a larger role than daily commuting. The authors said broader wastewater surveillance, paired with infection-control measures, could help slow early spread in future outbreaks.
Scientists at Columbia University’s Mailman School of Public Health used computer modeling to reconstruct how two major respiratory pandemics spread geographically across the United States: the 2009 H1N1 influenza pandemic and the 2020 COVID-19 pandemic.
The study—published in the Proceedings of the National Academy of Sciences—reports that in the United States, the 2009 H1N1 pandemic was linked to an estimated 274,304 hospitalizations and 12,469 deaths. For COVID-19, the researchers cited 1.2 million confirmed deaths reported so far in the United States.
To examine how quickly the pandemics moved between cities, the team combined detailed information about each virus’s transmission dynamics with simulations that incorporated air-travel patterns, everyday commuting flows, and the potential for superspreading events. The analysis focused on more than 300 U.S. metropolitan areas.
The simulations indicated that both pandemics established widespread circulation in most metro areas within a few weeks, often before early case detection or government response measures were in place. While the two viruses followed different pathways between locations, both relied on shared transmission hubs—including major metro areas such as New York and Atlanta—and the models found that air travel was more strongly associated with early spatial spread than commuting.
Unpredictable transmission patterns also introduced uncertainty that, the researchers said, makes real-time forecasting difficult. “The rapid and uncertain spread of the 2009 H1N1 flu and 2020 COVID-19 pandemics underscores the challenges for timely detection and control,” said the study’s senior author, Sen Pei, an assistant professor of environmental health sciences at Columbia. “Expanding wastewater surveillance coverage coupled with effective infection control could potentially slow the initial spread of future pandemics.”
The work also describes a broader modeling framework intended to be applied to other outbreaks. In addition to human travel patterns, the researchers noted that epidemic dynamics can be shaped by factors such as population demographics, school calendars, winter holidays, and weather conditions.
The study’s first author is Renquan Zhang of Dalian University of Technology. The author team includes researchers from Columbia University, Princeton University, and the U.S. National Institutes of Health, among others. The Columbia summary also notes that Jeffrey Shaman and colleagues have worked for more than a decade on methods to track and simulate infectious-disease spread to support public health decision-making.
