Decision Making

Interpretation of diagnostic test results is usu­ally part of the decision making process related to specific purposes such as those specified in Section 20.2. It is therefore difficult to have a single recommendation for each test result.

20.7.1 Certification

Animal herds cannot be certified free of MAP infection if insufficient numbers of animals are tested or if the animals tested are too young. The number of animals that would need to be tested depends on the Se and Sp of the specific antibody ELISA used, in addition to the herd demography. For example, Sergeant et al. (2008) suggested that a minimum of 80 adult animals should be tested to suggest that a herd has a low (but not zero) prevalence. An approach to correct for false-positive reactions caused by antibodies not specific to MAP was also suggested. Follow-up testing of ELISA­positive animals using a microbiological test cannot be recommended due to the frequency of ELISA-positive animals not shedding MAP (see Section 20.6.2).

The probability that a herd can be certi­fied as free of MAP-infected animals (P_F ) de­pends on the number of animals tested. P_F indicates that a herd has a lower prevalence than a predetermined prevalence, usually re­ferred to as the design prevalence. Herds with a sufficient number of animals tested and with a high P_lree can be classified as ‘potential­ly free of MAP infection'. A requirement for certifying herds ‘free of MAP infection' is that the herd is closed, or the animals purchased originate from ‘free' herds. Irrespective that a herd is classified as ‘potentially free', herds and animals are difficult to verify as free of MAP infection. Therefore, different levels of ‘freedom' can be used, e.g.

A special case of ‘certification' could be to determine whether an action plan has been ef­fective. An action plan to reduce transmission of MAP is pivotal in controlling MAP infections in general. Testing using a cell-mediated diag­nostic test such as IFN-γ in animals >15 months of age could be used to determine whether or not an action plan is working. If there are test­positive animals born after the action plan has been established, and the test is considered 100% specific, then the action plan should be revised because test-positive reactions would suggest that the animals have been exposed to MAP.

20.7.2 Reduction in transmission

An action plan to reduce transmission could include testing using antibody ELISA. Animals that are test-positive have a high risk of either being or becoming MAP infectious (lig. 20.2). Therefore, these animals should either be culled or measures to avoid transmission of MAP to susceptible animals should be established. Animals that are repeatedly negative in anti­body ELISA tests generally have a low probabil­ity of shedding MAP (Fig. 20.2). However, there is a chance that these animals might still shed MAP, particularly if the ELISA used has a low Se for detection of MAP-infectious animals. An increased test frequency can increase the overall probability of detecting infectious animals prior to the start of high bacterial shedding. According to Kudahl et al. (2008), a ‘sufficient' proportion of the infectious animals can be detected in this way, but it is unlikely that all infectious animals will be detected. Test-negative animals should be tested repeatedly, because they might become MAP-infectious at some stage in their life.

20.7.3 Increase in production parameters and animal welfare

Production parameters can be evaluated on ei­ther a herd or animal level. Consequently, pro­ductivity can be related to both the individual and the herd. Decisions relating to the individual would usually involve culling prior to the animal becoming affected by MAP infection, whereas decisions relating to the herd could also include decisions leading to a reduction in transmission (see above). At cow level, single-antibody ELISA results may be insufficient to determine whether a cow is MAP-affected. The test information must be combined with production data or clinical observations of the cow. A positive ELISA result in combination with diarrhoea or a decline in milk production should lead to immediate cull­ing to avoid further production losses. Animals with production loss or diarrhoea potentially related to MAP infection can be confirmed using ELISA.

Box 20.1. Summary recommendations.

An immune-based diagnostic test should be evaluated for its ability to detect MAP-infected and MAP- infectious animals in the population in which it will be used.

The purpose of testing should be clear, and the test results should be interpreted in relation to the purpose. In particular, special consideration should be given to the definition and communication of results, which could be ‘false positive' for one purpose and ‘true positive' for other purposes.

Cell-mediated immunodiagnostics may be used to determine whether a population has been ex­posed to MAP but cannot be used to take actions on an individual animal.

Humoral immunodiagnostics can be used to establish relative prevalence estimates, which can be used to compare estimates obtained previously with the same test.

Antibody ELISAs can only be used for certification if a sufficient number of animals are tested. A test-positive animal should not be confirmed by an agent-detecting test in a certification scheme. False-positive reactions can be excluded using mathematical formulas or retesting with other immune-based tests.

Antibody ELISA can be used for the risk-based management of MAP-infectious animals, since it will generally detect MAP-infected animals before they become MAP-infectious. However, frequent testing is a prerequisite if this approach is used.

20.8