The non-diseased reactor: considerations on the interpretation of screening test results.

The mathematical relationships between the sensitivity and specificity of screening tests and the proportion of animals which are positive to the test, but which have no visible lesions (NVL) at subsequent examination or necropsy, are outlined briefly. NVL reactors may be defined as a proportion of the total number of reactors to the test (reactor NVL per cent) or of the total number of animals tested (herd NVL per cent). Neither the herd nor reactor NVL per cent can provide valid estimates of the efficacy of any particular test, unless the precise prevalence of disease in the tested herd is determined. However under conditions of low or moderate prevalence specificity is approximately equal in magnitude to 100-herd NVL per cent but sensitivity cannot be reliably estimated from either the herd or reactor NVL per cent.