Current anti-hepatitis C virus (HCV) antibody screening immunoassays are routinely based on an indirect format. Although their use for anti-HCV antibody detection has achieved a very high specificity and sensitivity, false-positive results are still a problem especially among populations with a low prevalence of HCV infection. One strategy to obviate this problem is to adapt the assay from an indirect format to a double-antigen sandwich one to further improve its specificity. In this study, a double-antigen sandwich time-resolved immunofluorometric assay (DAS-TRIFMA) has been developed to detect total anti-HCV antibodies based on biotin-streptavidin interaction. For comparison, 1025 samples were analysed by the DAS-TRIFMA and three indirect anti-HCV antibody detection methods. For samples with discordant results, PCR-ELISA and Inno-LIA were employed as supplementary assays to analyse the presence of HCV antibodies. With regard to the 1025 clinical samples, the overall concordance between the DAS-TRIFMA and the three indirect methods was 99.41, 98.93 and 98.93 % for Ortho ELISA 3.0, WAT ELISA and I-TRIFMA, respectively. The specificity/sensitivity of the DAS-TRIFMA, Ortho HCV ELISA 3.0, WAT HCV ELISA and I-TRIFMA were 100/99.09, 99.34/98.18, 99.23/97.27 and 99.01 %/98.18 %, respectively. The DAS-TRIFMA was able to detect HCV antibodies at a concentration about 1/10 of that detectable by indirect methods. From the obtained results and their comparison, it is concluded that the DAS-TRIFMA is a more specific and reliable method for screening anti-HCV antibodies, and weakly positive S/Co values by the DAS-TRIFMA were more predictive of HCV infection than those by indirect methods.