Skip Over Navigation Links

Web Exclusives: Diseases

First Look: Policy Informatics
Emily Carlson and Stephanie Dutchen
Posted August 24, 2009

What happens when you let public health planners play with computer models? You can get faster, more effective measures to protect our well-being.

Exclamation iconThe field of policy informatics is so new that it doesn't have a Wikipedia entry yet.

This work is part of a new field called policy informatics that merges computation and decisionmaking. Right now, it's helping health officials around the country determine the best way to combat the H1N1 "swine" flu.

After the first cases broke out in April, the U.S. government called on disease modeler Stephen Eubank to help answer key questions about possible interventions, such as whether closing one school or all schools in New York City could stop the spread of flu from a few students to many city residents.

Ordinarily, Eubank and his team at the Virginia Bioinformatics Institute at Virginia Tech would tackle the questions: They'd use their sophisticated computational models to run simulations and get the answers. In this case, Eubank's team helped the public health analysts do it themselves.

Stephen Eubank. Credit: John McCormick, Virginia Tech Photo Department
Stephen Eubank talks about policy informatics, H1N1 and his modeling efforts. (Transcript)
Credit: John McCormick, Virginia Tech Photo Department

They retooled their own models to create an interactive system that was easy to use. The major challenge, Eubank says, was balancing the scientific rigor of modeling with the immediate deadlines of policymaking.

"Modelers inhabit this culture where there's plenty of time to think about issues, whereas analysts come from a culture where they need an answer now," Eubank explains.

Using the new system, analysts can log in online and, just like Eubank would have done, simulate a range of intervention strategies. The experiments take about a day to run, so analysts can think of a question on Monday and have results to discuss on Tuesday. If new questions or data emerge, they can log back in and run more simulations. The results are swift and sound.

"This kind of fast-turnaround modeling hasn't really been available to people making decisions during past outbreaks," says Eubank.

Eubank's work in policy informatics now will be aided by a $785,000 Recovery Act External link grant to extend our understanding of disease models and expand their use.

Listen to Eubank talk about policy informatics, H1N1 and his modeling efforts. (Transcript)

Learn about related research

This page last reviewed on April 22, 2011