Skip to main content

Highlight

Assessing the Relative Risk of RVF Introduction to the USA via Airline Traffic

Achievement/Results

The Quantitative Spatial Ecology, Evolution and Ecology (QSE3) IGERT Program at the University of Florida in Gainesville brings students together from multiple disciplines and departments to foster collaboration on and understanding of problems that span traditional academic boundaries. Each cohort of five trainees, comprising students from a range of disciplines, work together in an intense, one-year workshop on a project developed through partnerships with external clients, such as federal agencies or NGOS. The participating trainees are generally third- year students, having completed interdisciplinary coursework, an interdisciplinary rotation, and a few years of colloquium (a weekly one-credit course when all trainees and associates meet and discuss interdisciplinary research topics), as preparation for their workshop year.

Our fourth cohort of five trainees, Vincent Cannataro (Biology), Jake Ferguson (Biology), Andy Garcia (Geography), Elizabeth Hamman (Biology), and Jessica Langebrake (Math) are currently working on the risk of introduction of Rift Valley Fever into the US in collaboration with Seth Britch and Ken Linthicum of the USDA. Co-PIs Craig Osenberg (Biology) and Maia Martcheva (Math) have been overseeing the work of the students. Rift Valley Fever (RVF) is a mosquito-borne illness that can infect humans, wildlife, and livestock with serious, detrimental consequences. RVF was first discovered in Kenya in 1931 and consequently has spread throughout the African continent and beyond. A severe outbreak in Eqypt in 1977 caused thousands of cases of human illness and hundreds of deaths. RVF causes spontaneous abortion in livestock and could devastate the industry. There is some indication that RVF could spread across continents into Europe and potentially North America, highlighting the importance of quantifying the risk and identifying cost effective means of identifying and minimizing potential introduction. Previous researchers have quantified the risk of outbreaks in Africa using human and livestock densities, and rainfall to model mosquito population levels.

The students were able to incorporate these results along with data on the distribution of humans in Africa and global airline network flows, into a mathematical model of epidemiological dynamics know as the SIS model. SIS models incorporate the contacts between susceptible and infected individuals for a disease for which no immunity is conferred by having contracted the disease the first time. Using this model, the students were able to mathematically quantify the risk that infected individuals depart from an international airport in Africa and arrive at an international airport in the U.S. Preliminary results from the mathematical modeling indicate that JFK Airport, NY, has the highest risk of importation of RVF, with Atlanta and Dulles having high risk as well. The students are currently working on the first publication to come out of this work based on this study of airline data and risk.

Extending this work, the students are working on a sampling protocol for implementation at the at-risk international airports in the U.S. that would be cost-efficient but also effective at identifying infected individuals. Early detection upon arrival in this country is crucial for control of the outbreak; however to date, most theoretical studies have focused on containing outbreaks after they occur rather than determining efficient and effective sampling schemes to assess if the disease has entered a new environment. Sampling schemes are particularly challenging because many pathogens exist in both a host and a vector. The cohort therefore investigated how to sample with the goal of minimizing the time and cost of disease detection in a novel environment (such as Florida).

Preliminary results indicate that there is a critical time threshold in the course of the epidemic that determines the ideal sampling protocol. Above the threshold, the best sampling strategy is to sample both host and vector populations, while below the threshold it is better to focus on one or the other, but not both. These results provide practical advice for designing disease surveillance protocols. The students presented their preliminary results at the Epidemiology Research Institute Research Day (February 14, 2013). The workshop students, led by Vincent Cannataro, also produced a video that described their work for the 2013 IGERT poster and video competition. Their video explained their work in terms amenable to general audiences, and provided context for the importance of the work that they are doing. Their video won a public’s choice award and can be found at the url at the beginning of this section. The students are adding tourist behavior into the model — such as densities of tourists among affected African countries and the characterization of tourist destinations, such as rural or urban — to further refine the results of the model. Our 4th cohort of students has three more months of intense work on this project but will continue to work on the project over the coming year.

Address Goals

The intensive one-year workshop that each of the QSE3 IGERT cohort of five students takes part in is the capstone of the program. The students prepare for their workshop year by taking at least two interdisciplinary courses, spending a semester in rotation working with a professor outside of their home department and discipline, and participating in the 1-credit colloquium each semester where students learn about a variety of topics that span the disciplines that comprise the IGERT, including biology, geography, math, statistics, wildlife, and fisheries. The students are well prepared as an interdisciplinary group to work together to tackle a problem posed by an external client (though ultimately chosen by the students). The problem is addressed through an interdisciplinary approach with several related goals: 1) to resolve the problem for the client; 2) to contribute to basic knowledge that goes beyond the specific application defined by the client; and 3) to collaborate with scientists and non-academic clients with expertise outside of their core discipline.