Table 2 Main surveillance parameters with current estimates
Parameter definition | Symbol | Estimate (Range) | S-T range | Source / Derivation |
|---|---|---|---|---|
Maximum probability of detecting one infected animal in a herd test using the SICCT skin test | a | 0.48 (0.36–0.63) | 0.45–0.53 | In literature [22, 33] and estimated by fitting. The number of breakdowns disclosed by slaughterhouse surveillance is sensitive to this parameter. |
Maximum probability of detecting one infected animal when it is moved from a holding within an area where pre-movement testing is applied. | a m | 0.32 (0.24–0.42) | N | Detection reduced by exemptions from pre-movement testing. The recorded number of pre-movement skin tests was about ½ the number of movements from areas with pre-movement testing. Detection > 0.48/2 because infected cattle were more likely to be in movements without exemptions. |
Infection duration until SICCT reaches maximum sensitivity | t L | 35 (20–200) d | N | Experimental evidence [20]. |
Specificity of an individual SICCT skin test. | sp Sk | 0.9998 | N | The estimate of Goodchild et al. [34]. |
Proportion of age class i that is tested in a regular herd test | w 1 w2..w5 | 0 0.5–1.0 | 0 0.50–0.78 | Fitted to observed numbers of regular herd tests. |
Frequency of generic herd tests on infected herds | g i | Location dependent | 0.020–0.09 mth−1 | Fitted to observed numbers of breakdowns detected by generic herd tests. |
Frequency of generic herd tests on uninfected herds | g u | Location dependent | 0.0012–0.04 mth−1 | Fitted to observed numbers of generic herd tests. |
Maximum probability of detecting an infected animal at slaughterhouse surveillance. | a s | 0.875 (0.70–1.00) | 0.557–0.875 | Estimate may be high because of factors and practices absent from the model, including approved finishing units and age effects [35, 36]. Sensitivity increased during 2008–2010. |
Maximum probability of confirming an infected reactor. | a c | 0.40 | N | Fitted to number of OTFW (confirmed) breakdowns. This value underestimates the number of OTFW animals. (OTFW animals are unevenly distributed among breakdowns.) |
Time required for lesions to reach maximum detectability. | t Ls | 35 d (20–200) d | N | This and other models seem to fit using a wide range of values for related parameters [22]. Reports that a high proportion of young reactors have visible lesions indicate a low value [36]. |