Nathan Eagle is an expert in how to use cell phone data to predict and influence human behavior.
And he thinks that cell phones could be the next big thing in public health.
That’s because cell phone data could shed light on behavior connected to health issues. Researchers could potentially predict disease outbreaks by studying cell service providers’ data banks, which log records of all calls and text messages, including the cell phone tower that routed them. This information can enable researchers to measure mobility, helping them make inferences as to where people might be likely to travel—and thus how disease might be spread from location to location. Health care advice could be shared via cell phone. And public health researchers could collect crucial health data through cell phone surveys.
Eagle spoke to an HSPH audience on January 30, 2012 as part of the Infectious Disease Epidemiology Seminar Series.
An adjunct assistant professor in the Department of Epidemiology at Harvard School of Public Health (HSPH), Eagle holds academic positions at Northeastern’s College of Computer and Information Science and at the MIT Media Lab, where he earned his doctorate in “reality mining”—the gathering of data from cell phones or similar devices that relates to human social behavior. He’s also CEO of Jana, a company that provides technology that enables organizations to engage directly, through cell phones, with people in more than 85 countries.
The idea of using cell phones to improve public health is compelling, Eagle said. With 6 billion cell phone subscribers worldwide (out of the total world population of 7 billion), and cell phone penetration continuing to rise in growing or developing nations such as Nigeria, Indonesia, and Brazil, researchers can “start using the ubiquity of mobile phones to start actively improving the lives of people living in underserved, understudied regions around the world,” he said.
Incentives to communicate
While teaching and researching in rural Kenya several years ago, Eagle noticed that blood supplies often ran low at the local hospital. So he developed a system that compensated nurses—with cell phone airtime—for sending text messages about supply levels to the central blood bank.
That experience led him to form his company, which provides similar systems for a varied roster of organizations, from the United Nations to Procter & Gamble. Compensation in the form of airtime—valuable currency in many countries—can provide incentives for people to answer short surveys about everything from which soap they use to how often they visit a health clinic.
Tracking disease and health habits by analyzing behavior
One of Eagle’s research collaborations—with his wife, HSPH assistant professor Caroline Buckee—involves the use of cell phone data to help predict the potential spread of malaria. The researchers analyze the time and duration of calls, where calls are placed from, and where they’re connecting to. From this, they can draw inferences about future travel patterns of those making the calls—whether people from regions with high disease rates might travel to regions with lower rates or vice versa. This information could then guide public health efforts.
Eagle is also involved in a study analyzing the behavior of people who recently quit smoking in New York. The researchers are examining whether certain habits—like going to bars, hanging out with other smokers, or going out on Friday nights—are associated with relapse. They do this by examining behavioral data recorded by special smartphones given to the smokers in the study. “We want to build models that can infer the probability of someone deciding to have a cigarette based on their current context,” he said. “If we can learn the behavioral signatures associated with relapse, this could enable automated intervention.”
Taking the pulse of the ‘next billion’ mobile phone owners
Eagle is particularly excited by the prospect of active engagement with cell phone users the world over—a major step beyond simply gathering cell phone data about mobility or frequency of calls or texts. Through short cell phone surveys, he said, organizations could “take the pulse of the ‘next billion’ ”—mostly people in developing countries such as Africa and Asia—asking about everything from consumer products to health habits.
“Instead of just passive observation about people,” Eagle said, “I’m interested in a mechanism that lets us start engaging these populations, giving them questions they can respond to and rewarding them for particular behaviors.
--Karen Feldscher Photo: Photoshare