Zika Virus Speed and Direction: Reconstructing Zika Introduction in Brazil


  • Kate Zinszer Boston Children’s Hospital, Boston, MA, USA
  • Kathryn Morrison McGill University, Montreal, QC, Canada
  • John S. Brownstein Boston Children’s Hospital, Boston, MA, USA; Harvard Medical School, Boston, MA, USA
  • Fatima Marinho Ministry of Health, Brasilia, Brazil
  • Santos F. Alexandre Ministry of Health, Brasilia, Brazil
  • Elaine O. Nsoesie University of Washington, Seattle, WA




ObjectiveTo estimate the velocity of Zika virus disease spread in Brazil usingdata on confirmed Zika virus disease cases at the municipal-level.IntroductionLocal transmission of Zika virus has been confirmed in67 countries worldwide and in 46 countries or territories in theAmericas (1,2). On February 1, 2016 the World Health Organizationdeclared a Public Health Emergency of International Concern due tothe increase in microcephaly cases and other neurological disordersreported in Brazil (2). Several countries issued travel warnings forpregnant women travelling to Zika-affected countries with Brazil,Colombia, Ecuador, and El Salvador advising against pregnancy(3-7). The risk of local transmission in unaffected regions is unknownbut potentially significant where competent Zika vectors are present(8) and also given the additional complexities of sexual transmissionand population mobility (9,10). Despite the rapid spread of Zikavirus across the Americas and global concerns regarding its effectson fetuses, little is known about the pattern of spread. Knowledge ofthe direction and the speed of movement of disease is invaluable forpublic health response planning, including the timing and placementof interventions.MethodsData for this analysis were obtained from the Brazil Ministryof Health and consisted of confirmed cases of Zika virus disease.The centroids of the municipalities were taken in meters from theshapefiles and used to perform a surface trend analysis. Surfacetrend is a spatial interpolation method used to estimate continuoussurfaces from point data. The continuous surface of time to infectionwas estimated by regressing it against the X and Y coordinates. Timewas in days and X and Y coordinates were meters. Parameters wereestimated using least squares regression and velocity (in km per day)was obtained by inverting the final magnitude of the slope.ResultsData provided from the Brazil Ministry of Health on May 31,2016, indicated that Zika had been confirmed in 316 of the 5,564municipalities in Brazil representing 26 states, with six additionalmunicipalities identified from other reporting sources. Our modelsindicated a southward pattern of introduction of Zika starting fromthe northeast coast towards the southeastern coastal states of Rio deJanerio, Espírito Santo, and São Paulo. There was also a pattern ofwestern movement towards Bolivia. Overall, the average speed ofdiffusion was 42.1 km/day across all models was 6.9 km/day to amaximum of 634.1 km/day. The municipalities in the Northeast andNorth regions had the slowest speeds whereas the municipalities inthe Central-West and Southeast regions had the highest speeds. Thisis due to proximity of cases in time and space, with more cases havingoccurred closer in time and over larger areas in South, Southeast, andCentral-West regions resulting in faster rates of introduction.ConclusionsThe average speed of spread was 42 km per day and it tookapproximately five to six months for Zika to spread from thenortheastern coast to the southeastern coast and western border ofBrazil. The rapid spread of Zika can help us understand its possiblefuture directions and the pace at which it travels, which are key fortargeted mosquito control interventions, public health messaging, andtravel advisories. A multi-country analysis is needed to understand thecontinental spatial and temporal patterns of dispersion of Zika virus.




How to Cite

Zinszer, K., Morrison, K., Brownstein, J. S., Marinho, F., Alexandre, S. F., & Nsoesie, E. O. (2017). Zika Virus Speed and Direction: Reconstructing Zika Introduction in Brazil. Online Journal of Public Health Informatics, 9(1). https://doi.org/10.5210/ojphi.v9i1.7709



Communicable Disease Surveillance Use Cases for Human, Animal, and Zoonotic Diseases