Table 1. Description of the four datasets.
| Id | Data type | Number of herds (animals) | Period(geographic origin) | Control program | |
| Dairy | Beef | ||||
| A | Aggregated | 5 (134) | 1981–1983 (Nord) | Annual screening by SITT, slaughter of all cattle if>40% of SITT-positive animals, otherwise slaughter of SITT-positive only | |
| B | Individual | 1 (60) | 12 (625) | 2004–2006 (Dordogne) | Biennial screening by SITT confirmed by SICCT, slaughter of SICCT-positive animals, slaughter of all cattle in case of lesions and M. bovis isolation |
| Aggregated | 9 (684) | 16 (1,011) | |||
| C | Individual | 3 (142) | 11 (596) | 2007–2010 (Dordogne) | Annual screening by SITT with second control by γIFN1 and confirmation by SICCT, slaughter of SICCT-positive, slaughter of all cattle in case of lesions and M. bovis isolation |
| Aggregated | 7 (530) | 22 (1,344) | |||
| D | Aggregated | 29 (6,317) | 2005–2009 (Côte d’Or) | Biennial screening by SICCT, slaughter of SICCT-positive animals, slaughter of all cattle in case of lesions and M. bovis isolation | |
1
To improve the specificity of the control program C, the γIFN assay used in this control program is a combination of two tests: the PPD γIFN and the ESAT-6 γIFN. The modified γIFN test can be positive (both tests are positive), negative (both tests are negative) or divergent (the results of the two tests are discordant).