06:05 pm
23 October 2016

How Cell phone Data in Kenya could predict Rubella Outbreak

A team of data researchers at Princeton and other schools has announced that they can track disease outbreaks using cell phone data. The study published in the Proceedings of the National Academy of Sciences looked at the call histories of 15 million people in Kenya and correlated their call patterns with outbreaks of rubella throughout the year.

The paper was only possible because cell phone ownership has become so pervasive in the developing country. Outbreaks were shown to be more likely where kids congregate, namely schools. According to the study, calling patterns can be correlated with breakouts, vaccinations, and school closings.

Peaks showed in September 2008 and June 2009, also peaks in breakouts among Kenyan children. Spikes were also present from February-March 2009, but the cell phone data for February was not available for the study. Researchers were able to track locations and times, and map out callers’ movements according to those locations.

Why rubella was a perfect test case for predicting disease outbreaks according to cell phone data

Lead author C. Jessica Metcalf said rubella was a perfect disease to target in a data-driven test like this because of its dependency on human interaction to spread. “One of the unique opportunities of mobile phone data is the ability to understand how travel patterns change over time,” said Metcalf, who is an assistant professor at Princeton’s Woodrow Wilson School of Public and International Affairs.

Maps provided by a Princeton press release show call volume correlating with breakouts, but it was not immediately clear if the maps were significant because they also correlate with Kenya’s population density. Overall, the project segmented over 12 billion phone calls over the course of the year.

Photo courtesy of Amy Wesolowski, Harvard T.H. Chan School of Public Health; and C. Jessica Metcalf, Department of Ecology and Evolutionary Biology and Woodrow Wilson School of Public and International Affairs, Princeton University

“The potential of mobile phone data for quantifying mobility patterns has only been appreciated in the last few years, with methods pioneered by authors on this paper,” said lead author Amy Wesolowski, a postdoctoral fellow atHarvard’s School of Public Health in a press release.

This is not the first research to look into using cell phone data to follow disease outbreaks. Professor Alex Pentland at MIT’s Human Dynamics Laboratory and his student Anmol Madan, who together founded Ginger.io. Pentland in his book Social Physics describes the use of sensor data through smartphones as well as back-tracing callers’ movements to indicate where people would more likely catch the flu.

Metcalf describes herself as a demographer with an interest in “evolutionary ecology and infectious disease dynamics.” Coauthors hailed from Princeton, Harvard, the University of Southampton, the University of Pennsylvania, the University of Kabianga in Kenya, and Johns Hopkins Bloomberg School of Public Health. The study was supported by the National Science Foundation, the James S. McDonnell Foundation, the Bill and Melinda Gates Foundation and a number of other government agencies and private organizations.

Featured Image Credit: MEAACT Kenya / Flickr