Borna disease virus (BDV) infection of domestic animals and humans appears to have a worldwide distribution. There is evidence suggesting an association of BDV with certain psychiatric disorders. However, more comprehensive epidemiological studies are required to establish rigorously a link between BDV and human mental disorders, and to evaluate the role of carrier animals as potential source of BDV for human infection. The use of RT-PCR to detect BDV RNA in peripheral blood mononuclear cells (PBMCs) of infected individuals is a powerful tool to address these questions. The comparison of discrepant results reported by different investigators using this approach is hampered by the lack of controls to assess the sensitivity and reproducibility of the assays. Procedures are now described that allow the establishment of standardized controls to evaluate the performance of the RT-PCR assays. This RT-PCR assay detected reproducibly 100 copies of BDV p40 RNA in 5 microg of RNA. The data illustrate that the number of PBMCs used for RNA preparation, rather than the amount of RNA, has a critical influence on the outcome of the RT-PCR assay. Evidence is provided that levels of BDV in blood do not necessarily reflect viral load in brain.