Contagion as a Common Thread

Seth Richards-Shubik finds connections where others might see random topics.

Take, for example, adolescent sexual behavior, the European debt crisis and how physicians prescribe a new drug.

Richards-Shubik, an assistant professor of economics who studies economic networks in both health and finance, has examined "contagion" in all three areas.

In economics, contagion occurs when a shock in a particular area spreads out and affects others. Contagion can create positive or negative reverberations. Think economic crisis. Or economic boom.

For his thesis, Richards-Shubik studied contagion in adolescent sexual behavior. That led him to want to study contagion in other important areas such as the European debt crisis and new drugs.

"These are disparate topics, but the common thread is the methodology," says Richards-Shubik. "Contagion is the common thread."

In his research on physicians and a new drug for patients with atrial fibrillation, he and his colleagues found contagion strongly at play. But in his research on the European debt crisis, he and his collaborators found very little contagion.

His health care research focused on physicians in Pennsylvania and dabigatran, a blood thinner introduced in October 2010 that helps reduce patients' risk of stroke. It is sold under the brand name Pradaxa. Working with primary investigator Julie Donohue from the University of Pittsburgh, Richards-Shubik and several co-investigators discovered in preliminary results that physicians who treat the same patients are strongly influenced by their peers when prescribing a new drug. Their research is funded by the National Heart, Lung and Blood Institute of the National Institutes of Health.

"A common thread of the work is: Why do we see such disparity in treatment decisions in the same area? One potential reason is contagion among physicians," Richards-Shubik says. "First, you need to remember that medicine is not easy. Also, you need to realize that physicians learn from each other. There is this local transmission of knowledge. Idiosyncrasies can spread."

There are beneficial and harmful forms of contagion, he says. An influential doctor might choose to prescribe a cheaper, effective new drug, leading his colleagues to follow suit. Or he could decide to stay with a costlier, brand name drug even though a new, viable generic is available.

On the financial front, Richards-Shubik collaborated with Brent Glover, an assistant professor of finance at Carnegie Mellon University, to study contagion in the European sovereign debt crisis.

They examined 13 European countries and their financial ties with each other from 2005 to 2011. They wanted to see whether the economic collapse of one country, such as the one in Greece, would have a ripple effect in countries with financial ties to that country.

"Think about it in terms of banks," Richards-Shubik explains. "Banks loan money to each other all of the time. When a bank goes under, other banks are affected. It's a very clear and direct kind of contagion. It's a contagion based on asset values."

So is the same true when sovereign countries have financial ties with other countries?

"We actually found very little spillovers, very little contagion," Richards-Shubik says. "If I don't pay you back, are you hurt in your ability to pay other people back? The research suggests it does not. I was surprised at how weak contagion was from this mechanism.

Other factors may be at play."

Richards-Shubik says one factor that may have a strong impact is investors' beliefs.

"Do they actually believe in the possibility of a sovereign defaulting?" he asks.

In the case of the new drug and physicians' peers, "What surprised me in a sort of gratifying fashion was that the empirical results have actually met what you'd expect from economic theory," Richards-Shubik says.

WHY IT MATTERS

Understanding "contagion" in economic activities can be crucial for predicting the spread of beneficial or harmful outcomes, and for evaluating possible interventions.

Story by Janice Piccotti