Zika virus (ZIKV) recently emerged in the Americas and was declared a Public Health Emergency of International Concern by the WHO. Though Zika cases have been reported since the 1940s, the true burden of Zika is unknown. Estimates of the transmission potential and global burden of infection from ZIKV are fundamental to the design and targeting of effective interventions, but cannot be readily inferred from surveillance data due to misclassification with similar diseases like dengue, underreporting, and high proportions of asymptomatic disease

We combined mathematical modeling and novel serological testing of existing blood samples from five countries around the world to estimate key transmission parameters and risk maps for ZIKV. While the primary focus of this project was ZIKV, the data generated and methods employed could also be applied to related arboviruses (such as dengue, chikungunya and Japanese encephalitis). The spatial overlap between ZIKV and similar diseases also provides key insight into the future spread of Zika and other viruses.

We first developed statistical techniques to better analyze data from the SHERPADES-Luminex assay, a novel test for past exposure to ZIKV and related viruses, using well characterized samples from Colombia and the Philippines. We tested over 5,000 existing serum samples using this assay. Initial results from Bangladesh, Burkina Faso and Gabon showed little or no evidence of ZIKV circulation. Results from the Philippines showed that around 8% of the population had been infected. Finally, results from Colombia showed closer to 55% ZIKV seropositivity. There is strong evidence of the circulation of other arboviruses across these locations. Using the results of the serological testing of samples from Burkina Faso, we fit mathematical models of the past circulation of different arboviruses in the region. In particular, we have been able to demonstrate that there have been four short-lived outbreaks of chikungunya in Burkina Faso over the past 60 years (Figure 1, attached). There is no previous report of chikungunya in the country. Further, we demonstrate the recent entry of chikungunya into Gabon, with no circulation prior to 2007.

Estimates of the reproductive number of Zika in counties in Florida in which local cases occurred ranged between 0.3 and 0.7, suggesting that on average each case resulted in 0.3 to 0.7 locally acquired cases.

We believe these results will be informative to public health decision makers when planning prevention and intervention strategies for Zika and related arboviral diseases. We have also partnered with local researchers in several countries to provide training in the SHERPADES-Luminex assay used in this study, and hope that it may be used as a surveillance and clinical tool moving forward.

Last Modified: 03/02/2018
Modified by: Derek A Cummings