NIH chooses Los Alamos to model urban epidemics

An emergency room physician sees a patient with a high fever and a trace of a rash and admits her to the hospital. The next morning, three more patients with similar symptoms come in, then more until lab tests confirm the initial hunch: an outbreak of smallpox has begun.

How to keep the outbreak from becoming an epidemic, and recommending the best responses to public health officials, could be revealed through computer simulations under development at the Laboratory.

The National Institute of General Medical Science, or NIGMS, today chose a team of scientists at Los Alamos to study how people's daily interactions might affect the spread of and efforts to stem infectious diseases, whether natural or deliberately released by bioterrorists.

The project is one of four studies of epidemics and community responses funded through the Models of Infectious Disease Agent Study, or MIDAS by the Institute, which is part of the National Institutes of Health. The projects also will help in biodefense, allowing officials to simulate strategies to detect, control and prevent the spread of disease. Information about the studies is available at http://www.nigms.nih.gov/news/ online.

"This grant recognizes years of development effort by many computer scientists and mathematicians at Los Alamos who have dedicated their careers to understanding social networks and their potential vulnerability to attack or natural disaster," said Bill Feiereisen of Computer and Computational Sciences (CCS-DO) Division.

The Los Alamos team, led by Stephen Eubank of Basic and Applied Simulation Science (CCS-5), will build an artificial city of 1.5 million people in a computer, then explore how patterns of human contact affect disease transmission, as well as the consequences of early and mid-term responses. The new project, "Population Mobility Models of Urban Disease Outbreak," is based on EpiSims, one component of a well-established family of Los Alamos epidemiological computer models known as the Urban Infrastructure Suite. The NIGMS awarded Los Alamos $3.6 million over the next five years.

The simulations developed through the project will allow public health officials or other users to specify such scenarios as the initial health of the population, how people interact with disease-causing pathogens, possible paths for introduction of the pathogens, and the choice, timing and location of strategies to fight the outbreak.

"Models such as this one are valuable tools in understanding population vulnerabilities, planning operations when disease breaks out and assessing the costs and potential benefits of mitigation strategies, including how they impact related infrastructures such as transportation, schools and local health care," Eubank explained. "They won't predict whether John or Jane Doe will get sick, but they might help prevent an epidemic from spreading to the Does' neighborhood."

At the heart of the simulations that Eubank's team will develop for NIGMS are methods to track the transmission of multiple interacting diseases in an artificial, but statistically accurate, representation of an urban population and a mathematical, structural analysis of contact and transmission patterns.

More generally, the Los Alamos team hopes to learn which features of social networks are most prominent in determining responses to epidemics and efforts to stop them. The models the team develops will evaluate these features, estimate how they vary in cities of different sizes and characteristics, and generate realistically random instances of social networks.

Eubank and his team will regularly pass on their findings and new simulations to the MIDAS Informatics Group and work with them on further development of analysis tools for social networks as well as useful epidemiological data based on likely disease scenarios.

The Los Alamos team and the other MIDAS scientists also will be able to change hats quickly, from theoreticians seeking a fundamental understanding of biological processes, to experimentalists helping government officials to make quick decisions about how to handle an actual epidemic.

-- Jim Danneskiold

Other Headlines