Synthetic population study offers new strategy for controlling epidemics in big cities
January 23, 2014
Researchers in the Network Dynamics and Simulation Science Laboratory at the Virginia Bioinformatics Institute are the first to model in detail how transient populations impact the spread of an illness, and how outbreaks such as influenza can be curbed by encouraging healthy behaviors in high-traffic tourist destinations.
Influenza places a huge burden upon society, both physically and economically. It is estimated that influenza costs the United States economy over $87 billion annually.
In a large city like Washington, D.C., with about 50,000 visitors on any given day who stay for just a few days, there is a constant influx of new people who are susceptible to infections. Further, they visit highly populated tourist destinations, where they come into contact with other visitors as well as residents. Disease can spread quickly.
“We built a detailed synthetic population model of Washington, D.C., including transient populations: tourists, business travelers," said Samarth Swarup, an applied computer scientist at the institute. "Our computational model shows that an influenza epidemic can be much worse when we take the impact of transients into account.”
Researchers used EpiSimdemics, an interaction-based, high-performance computing simulation software program to simulate the spread of a flu-like disease in the Washington, D.C., metro area.
The Network Dynamics and Simulation Science Laboratory integrates analytical and computational research across diverse domains to develop synthetic information systems and analytical methods to understand very large, complex systems. Investigators perform basic research and develop informatics technology.
The simulations determined whether closing the area’s main museums for varying durations would affect the spread of the illness, and also whether the spread of the illness was slowed if healthy behaviors, such as covering coughs and using hand sanitizers, were encouraged at these main tourist destinations.
They also analyzed the spread of the epidemic and derived the average number of contacts per day, per individual, and the average duration per contact.
Their studies revealed that by encouraging healthy behaviors at locations of high mixing, such as the museums of the Smithsonian Institution, the outbreak size could be significantly reduced, and the peak of the epidemic could be significantly reduced and delayed. It turns out that this is much better than simply closing the museums for a few days a kind of "social distancing" intervention, which seems to have no effect on the epidemic.
Researchers say the findings, which appeared in November in the journal Scientific Reports, can be used for policy recommendations, such as promoting the use of hand sanitizers in museums.
“That in turn would offer the opportunity to conduct a field experiment to validate our model against actual epidemic and intervention data,” Swarup said.
The research was supported in part by Defense Threat Reduction Agency contract HDTRA1-11-D-0016-0001, the National Institutes of Health and National Institute of General Medical Sciences Models of Infectious Disease Agent Study grant 2U01GM070694-09, and National Science Foundation grant SES-0729441.