[en] In order to contribute to the definition of the best ELISA test for screening and differential diagnosis of Brucella abortus to be applied for control programmes, a total of 2971 sera from Colombian cattle were tested for brucellosis. Conventional agglutination tests, Buffered Plate antigen test (BPAT) and Rose Bengal (RB) as well as Complement Fixation test (CFT) (Alton, et al. 1988) were used comparatively. Radial immunodiffusion test (RID) was also performed to all sera. The sera were also tested using four different ELISAs: indirect ELISA from FAO/IAEA and the indirect ELISA modified by Nielsen, et al. 1992 as well as two competitive ELISAs: one competitive ELISA used B. abortus O-polysaccharide antigen and an enzyme conjugated monoclonal to the O-polysaccharide for competition and detection. The second competitive ELISA used lipopolysaccharide (sLPS) antigen, a different monoclonal antibody for competition but also specific for the O-polysaccharide and a commercially available goat anti-mouse IgG enzyme conjugate for detection. The sera were analyzed based on its population status, 987 positive obtained from Brucella abortus infected herds based on clinical and/or bacteriological evidence and a high prevalence of brucellosis, CFT percentage of positive animals in the herd was greater than 5%. Eight hundred sixty six (866) negative sera from non-vaccinated cattle from a brucellosis free area and 1118 negative sera obtained from reglamentary vaccinated areas under a free herd program. Initial cut-off values were derived using negative serum samples. The diagnostic sensitivity and specificity was defined from frequency histograms based on this cut-off values and using 2x2 tables, corresponding confidence limits (95%) were calculated. The data were also analysed using signal detection analysis (ROC). Kappa statistics was determined for all tests and populations, accuracy was used as index of comparison to evaluate different assays. The data support the initial hypothesis that the ELISA methodology designed for brucellosis will provide more precise and standardised method for diagnosis and for the support of control and eradication campaigns. (author)