Studies of blood donors positive for antibody to hepatitis C virus on enzyme-linked immunosorbent assay are probably biased by the large number of false-positive results. We evaluated the epidemiological and biological characteristics of 177 such donors with regard to the confirmatory second-generation RIBA test (Ortho Diagnostic Systems) and have compared the results to those from an age- and sex-matched control group of 177 donors negative for antibody to hepatitis C virus on enzyme-linked immunosorbent assay. Second-generation recombinant immunoblot assay was positive in 38% of cases, indeterminate in 6% and negative in 56%. The case-control study showed a significantly higher frequency of intravenous drug abuse (27% vs. 0%), blood transfusion (22% vs. 9%), history of jaundice (21% vs. 7%), elevated ALT level (49% vs. 4%) and HBc antibody positivity (32% vs. 7%) in second-generation recombinant immunoblot assay-positive donors. No such differences were found between the second-generation recombinant immunoblot assay-negative donors and their controls. The 35 second-generation recombinant immunoblot assay-positive donors without histories of transfusion or intravenous drug abuse had a significantly higher frequency of surgery with major blood loss or prolonged stays in areas of hepatitis B virus endemicity than did their controls (20% vs. 0% and 49% vs. 26%, respectively). In conclusion, at least one risk factor for HCV infection was identified in 82% of the second-generation recombinant immunoblot assay-positive donors, 91% of whom could have been identified on the basis of these risk factors, ALT level and presence of HBc antibody.