Schlüsselwort: Geflügelpest

Retrospective analysis of biosecurity and the likelihood of wind-mediated transmission of highly pathogenic avian influenza viruses between turkey farms in north-west Germany During the epidemic of highly pathogenic avian influenza (HPAI) in Germany in 2016/2017, commercial turkey farms in the districts of Cloppenburg and Oldenburg in Lower Saxony were affected with striking frequency. The density of turkey farms is particularly high in these districts. Due to the conspicuous spatial and temporal accumulation of HPAI outbreaks, windmediated transmission of HPAI viruses between affected turkey farms was discussed as a possible cause of introduction. In this study, the biosecurity of the farms was investigated with regard to possible links to the disease outbreak and the possible HPAI virus spread by wind-mediated transmission between affected turkey farms in the districts of Cloppenburg and Oldenburg during the 2016/17 avian influenza epidemic analysed. Data recorded from all turkey farms registered within the epidemic period were processed and analysed. The data had been collected by the responsible veterinary officials with the help of the "Checklist for turkey biosecurity" of the German federal state of Lower Saxony. A univariable analysis was carried out, whereby the occurrence of HPAI on the farm was tested for an association with biosecurity-related variables. After checking for correlation, a multivariable analysis using logistic regression was performed. This identified aspects of farm management, which were associated with the incidence of avian influenza in the flocks, as potential risk factors. These relate to the type and frequency of health checks of the livestock and the use of farm vehicles on several farm sites and the number of stables per holding. Some of the identified risk factors in the biosecurity measures persisted after the end of the epidemic. To analyse the probability of wind-mediated transmission of HPAI viruses, each possible constellation between farms, consisting of the potential HPAI virus-emitting and the potential HPAI virus-receiving farm, was tested in several selection steps for spatial, temporal, virusgenetic and wind conditions. For each farm constellation, individual worst- and best-case scenarios were considered. These differed with regard to the assumed HPAI virus emission in the potentially HPAI-virus-emitting farm. All farm-specific risk days were examined, on which potentially HPAI-infected turkeys, their carcasses or potentially HPAI virus-contaminated litter mixture were present on the farms, which could have been the source of virus emission. In a worst-case scenario, maximum virus emission was assumed in all potentially HPAI virusemitting farms on all potential risk days, in a best-case scenario only for those farms that were classified as HPAI virus-emitting based on the case number analysis of the HPAI-positive flocks. As a result, wind-mediated transmission of HPAI virus was classified as "likely" in two (5.4%) of the 37 HPAI outbreaks and "conditionally likely" in nine (24.3%) HPAI outbreaks. The results allow the conclusion that wind-mediated transmission of HPAI virus between affected turkey holdings was possible in 29.3% of the cases in the area studied. The results of the study show that wind transmission can be ruled out as the sole cause of epidemiologically linked HPAI outbreaks in turkey farms in the districts of Cloppenburg and Oldenburg in 2016/2017. In contrast, certain deficits in the biosecurity systems were statistically significantly more frequently detectable in outbreak farms than in farms without an outbreak, even after the end of the epidemic.