Analysis of Estonian surveillance in wild boar suggests a decline in the incidence of African swine fever

Abstract

African swine fever (ASF) in wild boar populations is difficult to control. In affected areas, samples from all wild boar shot and found dead are investigated. The use of laboratory tests allows estimating the duration of the infection in affected animals. The study aimed to test the hypothesis that the stage of the epidemic in different areas of Estonia can be assessed on the basis of prevalence estimates. ASF surveillance data of Estonian wild boar were used to estimate prevalences and compare them between the East and West of Estonia. The temporal trend of the estimated prevalence of ASF virus positive animals and of the estimated seroprevalence of wild boar showing antibodies against ASFV was analyzed. Due to the potential influence of population density on the course of ASF in wild boar, also population density data (number of wild boar/km²) were used to investigate the relationship with laboratory test results. In areas, where the epidemic had already lasted for a long time, a small number of new cases emerged recently. The prevalence of samples that were only seropositive was significantly higher in these regions as compared to areas, where the epidemic is in full progress. The observed course of the disease could be the beginning of an ASF endemicity in this region. However, the results may also indicate that ASF has started to subside in the areas that were first affected in Estonia.

Figure 1

The study areas “East” and “West” in Estonia. The island of Hiiumaa was excluded. Map was generated by using ArcGIS ArcMap 10.3.1 (ESRI, Redlands, CA, USA).

Figure 2

Timescale illustrating the course of laboratory test results for African swine fever.

Figure 3

Median temporal effect of all samples from area “East” (E) and area “West” (W) that tested PCR-positive, irrespective of the serological result, on the logit prevalence. 95% Bayesian credible intervals (BCI) are indicated. Figure was generated by using the software package R (http://www.r-project.org).

Figure 4

Median temporal effect of all samples from area “East” (E) and area “West” (W) that tested exclusively serologically positive on the logit prevalence. 95% Bayesian credible intervals (BCI) are indicated. Figure was generated by using the software package R (http://www.r-project.org).

Figure 5

Wild boar population density (number of wild boar/km²) in the different hunting years for area “East” (E) and area “West” (W). The horizontal line that forms the top of the box illustrates the 75th percentile. The horizontal line that forms the bottom is the 25th percentile. The horizontal line that intersects the box is the median number of wild boar per square kilometer. Whiskers represent maximum and minimum values that are no more than 1.5 times the span of the interquartile range and the open circle represent outlier, which are single value greater or less than the extremes indicated by the whiskers. Figure was generated by using the software package R (http://www.r-project.org).