Diagnostic laboratories are an essential component of animal disease surveillance systems. To understand the occurrence of disease in populations, surveillance systems rely on random or targeted surveys using three approaches: clinical, serological and virological surveillance. Clinical surveillance is the basis for early detection of disease and is usually centered on the detection of syndromes and clinical findings requiring confirmation by diagnostic laboratories. Although most of the tests applied usually perform to an acceptable standard, several have not been properly validated in terms of their diagnostic sensitivity and specificity. Sensitivity and specificity estimates can vary according to local conditions and, ideally, should be determined by national laboratories where the tests are to be applied. The importance of sensitivity and specificity estimates in the design and interpretation of statistically based surveys and risk analysis is fundamental to establish appropriate disease control and prevention strategies. The World Organisation for Animal Health's (OIE) network of reference laboratories acts as centers of expertise for the diagnosis of OIE listed diseases and have a role in promoting the validation of OIE prescribed tests for international trade. This paper discusses the importance of the epidemiological evaluation of diagnostic tests and the role of the OIE Reference Laboratories and Collaborating Centres in this process.