Political and environmental factors – e.g., regional conflicts and global warming – increase large-scale migrations, posing extraordinary societal challenges to policy makers of destination countries. A common concern is that such a massive arrival of people – often from a country with a disrupted healthcare system – can increase the risk of vaccinepreventable diseases outbreaks like measles. We analyze human flows of 3.5M Syrian refugees in Turkey inferred from massive mobile phone data to verify this concern. We use multilayer modeling of interdependent social and epidemic dynamics to demonstrate that the risk of disease re-emergence in Turkey, the main host country, can be dramatically reduced by 75–90% when the mixing of Turkish and Syrian populations is high. Our results suggest that maximizing the dispersal of refugees in the recipient population contribute to impede the spread of sustained measles epidemics, rather than favoring it. Targeted vaccination campaigns and policies enhancing social integration of refugees are the most effective strategies to reduce epidemic risks for all citizens [1]. Transmission model. To model measles transmission in Turkey and the mobility of Turkish and Syrian refugees, we assume two populations of individuals, namely population T of size N(T) and population R of size N(R) , living in a territory consisting of L distinct geographically patches (i.e. Turkish prefectures) accounting for N (T) k and N (R) k individuals, k = 1, ..., L with P L k=1 N (T) k = N(T) and P L k=1 N (R) k = N(R) . The formulation of the force of infection for the two populations encodes how the interplay between the level of mixing of refugees with local populations and their mobility patterns shapes the spatial spread of simulated epidemics. Let us indicate by c (p) ki (p ∈ T, R) the elements of a matrix C(p) encoding the number of people belonging to population p travelling from patch k to patch i, and with α the fraction of Syrian contacts with Turkish citizens. The force of infection for each population in the i–th patch depends on the contribution of all patches in the country:
