BackgroundIn 2008, a program to eradicate bovine virus diarrhea (BVD) in cattle in Switzerland was initiated. After targeted elimination of persistently infected animals that represent the main virus reservoir, the absence of BVD is surveilled serologically since 2012. In view of steadily decreasing pestivirus seroprevalence in the cattle population, the susceptibility for (re-) infection by border disease (BD) virus mainly from small ruminants increases. Due to serological cross-reactivity of pestiviruses, serological surveillance of BVD by ELISA does not distinguish between BVD and BD virus as source of infection.ResultsIn this work the cross-serum neutralisation test (SNT) procedure was adapted to the epidemiological situation in Switzerland by the use of three pestiviruses, i.e., strains representing the subgenotype BVDV-1a, BVDV-1h and BDSwiss-a, for adequate differentiation between BVDV and BDV. Thereby the BDV-seroprevalence in seropositive cattle in Switzerland was determined for the first time. Out of 1,555 seropositive blood samples taken from cattle in the frame of the surveillance program, a total of 104 samples (6.7%) reacted with significantly higher titers against BDV than BVDV. These samples originated from 65 farms and encompassed 15 different cantons with the highest BDV-seroprevalence found in Central Switzerland. On the base of epidemiological information collected by questionnaire in case- and control farms, common housing of cattle and sheep was identified as the most significant risk factor for BDV infection in cattle by logistic regression.ConclusionThis indicates that pestiviruses from sheep should be considered as a source of infection of domestic cattle and might well impede serological BVD surveillance.Electronic supplementary materialThe online version of this article (doi:10.1186/s12917-016-0932-0) contains supplementary material, which is available to authorized users.