Abstract / Summary
Accurate diagnosis of Rift Valley Fever (RVF) disease is critical for outbreak surveillance, containment, intervention, trade, and public health. In this study, we sought to review available evidence on characteristics of validated RVF diagnostic assays used in livestock, humans and wildlife globally. We systematically searched five major databases (1931–2023) for studies reporting the accuracy of RVF diagnostic assays. Data were extracted on test performance, host species, geography and methodological quality using QUADAS-2. From 5,285 screened records, only twenty-eight studies met our eligibility criteria. 79% (22/28) focused on serological assays (IgG/IgM ELISA), with over 5,200 livestock samples intended to measure past exposure. In contrast, only six studies (21%) evaluated molecular tests (RT-PCR/LAMP), which detect current infection and are vital for outbreak detection. Molecular test evaluations are based on a limited total of fewer than 1,500 samples, with reported sensitivity ranging from 71% to 100%; however, this wide range reflects both genuine performance variation and the very small number of validation studies available. This disparity likely stems from practical and logistical challenges, including the high cost of multi-centre studies, the difficulties in obtaining well-characterised clinical samples, and the stringent biosafety requirements for handling live virus. Wildlife studies are virtually absent (only two in African buffalo). A final concern is that blinding of reference standards was unreported in 89% of studies, which could contribute to an overestimate of test performance. This review highlights the need to strengthen the evidence base for RVF diagnostics evaluation, particularly for molecular assays used in active surveillance. We call for increased investment in multi-centre validation studies, developing standardised reference materials, and enhancing laboratory capacity in endemic regions.