Battling Ebola with Cell Phone Data
By Joni Blecher
The ongoing outbreak of the Ebola virus in West Africa has many people thinking about better ways to contain and track the spread of the disease. One solution may be found in a device that billions of people around the world carry every day: the cell phone. It’s not as creepy as it may sound. We’re not talking about tracking every movement of a single phone, but rather following the movements of many phones throughout a given area to determine where a general population travels.
Whenever a call is placed using a cell phone, a call detail record (CDR) is generated. It includes information such as the phone numbers involved in the call, duration, type of communication (voice or text), and the cell tower that handled it. Carriers use CDRs to generate phone bills. However, when reviewing the data in a larger set, it can provide a lot more information about the activity of people in a specific area and reveal patterns. That’s how cell phone data can help build models of population movement—and potentially predict where a disease will spread next.
There have already been a few studies that used CDR data to track diseases. In 2006, Harvard epidemiologist Caroline Buckee and her husband Nathan Eagle (CEO of Jana, a company that conducts surveys via mobile phones) traveled to Africa. Together, they were able to analyze the data of calls and text messages in a certain region. They also followed where those phones went, and looked at the spread of malaria in those areas. They were then able to compare that data to satellite images of the region and pinpoint disease hot spots—providing valuable information about the spread of malaria and its origins. Armed with this type of intelligence, health workers could focus their efforts in the regions that would benefit most.
Flowminder Foundation, a non-profit organization based in Stockholm, is currently using a similar model to help identify national mobility estimates for West Africa to help monitor the current Ebola outbreak. By using data provided by the region’s cell phone service provider, Orange Telecom, these models can show where the disease has spread and where it might go next—enabling medical teams to start deploying preventive measures and hopefully contain the outbreak.
Being able to understand where people go during an epidemic or after a natural disaster has been helpful in properly administering aid during times of need. After the 2010 earthquake in Haiti, cell phone data was used to monitor population movement. Researchers from the Karolinska Institute in Sweden and Columbia University in New York used cell phone data from the country’s largest cell phone provider, Digicel, to track the daily travels of 2 million people.
Following the quake was an outbreak of cholera. The data helped researchers identify where people went after leaving an affected area, and thereby plot where the disease would spread. Even more compelling is that they were able to generate estimates within a mere 12 hours of receiving the data, which means aid workers could potentially save more lives by quickly redirecting their limited resources.
Although there are large quantities of CDRs being used to create these models, there are still privacy issues to consider. While researchers aren’t drilling down to granular levels of the data, there is personal information contained in each CDR—information that could be exploited in the wrong hands. Is it acceptable for cell phone service providers to hand over such detailed information, no matter how noble the cause? That’s an issue currently being worked out by researchers and the mobile industry’s global trade organization, the GSMA.
Joni Blecher is a freelance writer who has spent her career covering tech and a myriad of lifestyle topics. When she’s not writing, you can find her exploring the food scene in Portland, OR.