Diagnostic test evaluations (DTEs) for chronic infections are challenging because a protracted incubation period has to be considered in the design of the DTE, and the adverse effects of infection may be widespread and progressive over an animal's entire life. Frequently, the specific purpose of the test is not formally considered when a test is evaluated. Therefore, the result is often a DTE where test sensitivity and specificity estimates are biased, either because of problems with establishing the true infection status or because the test detects another aspect of the infection (and analyte) than originally intended.

The objective of this paper is to outline a structured approach to the design and conduct of a DTE for diagnostic tests used for chronic infections in animals, and intended for different purposes. We describe the process from reflections about test purpose and the underlying target condition through considerations of the pathogenesis, and specification of a practical case definition, which can subsequently be used in the DTE for the specific purpose.

The process is illustrated by two examples of Mycobacterium avium subsp. paratuberculosis (MAP) infections in cattle. MAP infections are chronic and can result in different adverse effects at different time points during the incubation period. The description provides input on the process and deductive reasoning which are integral parts to develop a high-quality design of a DTE for chronic infectious diseases