Skip to main content
ARS Home » Plains Area » Kerrville, Texas » Knipling-Bushland U.S. Livestock Insects Research Laboratory » LAPRU » Research » Publications at this Location » Publication #354132
Title: African swine fever research in Ukraine: Spatio-temporal epidemiological analysis and exploratory risk mapping of outbreaks

Author
item FILATOV, SERHII - National Scientific Center
item STEGNIY, BORYS - National Scientific Center
item GERILOVYCH, ANTON - National Scientific Center
item Perez De Leon, Adalberto - Beto

Submitted to: Meeting Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 9/7/2017
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: African swine fever (ASF) is a high-consequence viral disease that threatens pig production industry in Eastern Europe, including several EU member states. Ukraine is experiencing regular ASF outbreaks among domestic pigs and in wild boar populations since early 2014; however, no attempt to analyze the distributional patterns and determine high risk areas for disease spread in this country has been made thus far. We compiled a comprehensive georeferenced dataset of countrywide ASF case notifications derived from the national surveillance system and applied a retrospective space-time scan statistics to the data to detect hotspots of disease emergence. Furthermore, to predict high risk areas for ASF occurrence in the territory of Ukraine, we modelled disease distribution with Maxent algorithm using relevant open source GIS data as predictor variables. Eight significant (p<0.005) clusters of ASF cases with the mean duration of 148.75 (13-433) days and the mean radius of 93.01 (40.27-149.98) kilometres were detected using the space- time permutation model. Most of the detected clusters were located in transboundary regions, close in space and time to ASF cases reported in affected neighbouring countries. The Maxent models accurately identified areas at risk of ASF outbreaks with the average AUC of 0.8622 and 0.8104 for the training and test data respectively. The three average most important predictors were: Human population density (28.1%), Land Cover (13.2%), and Nightlight intensity (11.5%). These results suggest a substantial international border porosity in the high risk areas, which apparently facilitates the transboundary spread of ASF in domestic pig and wild boar populations. Future studies highlighting (dis)similarities in environmental features, animal husbandry, cultural practices, and smallholder pig value chain characteristics in these territories could prove useful to better understand the spread of ASF in this part of Europe.