Less Bite for Your Buck: Using Cell Phone Data to Target Disease Prevention∗

Infectious diseases have a large economic and social burden that is magnified by infected travelers who spread diseases to the locations they visit. In this paper, I study malaria transmission in Senegal by quantifying the relationship between travel and spread of the disease and showing its implications for targeting policies. Using individual mobile phone records for 9 million users in 2013, I estimate daily interregional movement and demonstrate substantial intertemporal and geographic variation in movement. I link this variation to clinic data on the incidence of malaria in order to calculate the probability a traveler is infected and to determine the impact in the area a traveler enters. Estimates indicate that an infected traveler entering a health facility’s catchment area causes reporting of 1.6 additional cases. I apply the results to evaluate the potential for policies targeting travelers. At the same cost, strategic targeting of travelers from high-incidence locations would result in up to five times as many cases being averted as compared to current policies of randomly targeting travelers during the malaria season. These findings indicate how novel applications of big data, combined with traditional health measures, can enable improvements in policy to address negative spillovers from travel and lower the burden from communicable disease. JEL Classification: J18, 021, I18, H51 ∗I am grateful to Andrew Foster, Jesse Shapiro, and Daniel Bjorkegren for their continuous guidance and support on this project. In addition, I would like to thank Andrew Tatem, Nick Ruktanonchai, Emily Oster, Bryce Steinberg, Victor Alegana, Morgan Hardy, David Glick, Girija Borker and William Violette for valuable feedback. I also appreciate the helpful comments from participants at the World Bank ABCDE Conference, the 2016 NEUDC Conference, the 2016 PopPov Conference, the 2016 Population Health Sciences Workshop, the PODER Summer School, the 2016 PAA Conference and the Brown microeconomics lunch seminar. Thank you to the Bill & Melinda Gates Foundation and the National Institute for Child Health and Human Development (Grant # T32 HD 7338-29) for financial support. Finally, this research would not have been possible without the data provided by Orange and Sonatel’s Data for Development Challenge, the National Malaria Control Program and MACEPA.