Interest in the assessment of autoantibody specificity stems from the need for an autoantibody marker capable of predicting clinical events in autoimmune disorders. However, the multiplicity of epitopes present on autoantigenic particles, the quantitative and qualitative heterogeneity of autoantibodies, as well as the nature of the tests, mean that each of the assays used in their determination have different characteristics. The aim of this study was to compare the specificities of different ANAs using four commercial assays. The routine method used for the detection of ANA is indirect immunofluorescence on Hep-2 cells. The assays used were: counterimmunoelectrophoresis (CIE), enzyme-linked immunosorbent assay (ELISA), and two immunoblotting assays. Kappa statistic was applied to evaluate the consistency between tests. Kappa index is a measure of agreement between categorical data. Kappa has a maximum of 1.00 when the agreement is perfect, a value of zero indicates no agreement better than chance, and negative values show worse than chance agreement. For SS-B antibodies, there was a good concordance between all four methods used (Kappa 0.66-0.74). For anti RNP antibodies, the results for CIE/ELISA (Kappa 0.60) were consistent as were the two immunoblot methods (Kappa 0.69). For anti Scl-70 (topoisomerase I) antibody, results from the ELISA and CIE methods were totally consistent (Kappa 1.00). In spite of the high prevalence of anti SS-A/Ro antibodies, the agreement between the methods was poor, without statistical significance. Finally, for Sm antibodies, more consistent results were obtained between CIE/ELISA (Kappa 0.51) and between one of the immunoblotting methods and ELISA (Kappa 0.54). In conclusion, CIE concurs mostly with ELISA for anti-RNP, Scl-70, Sm and SS-B antibodies, but with some disagreement for SS-A antibodies.