Implementation of routine blood donor screening using anti-hepatitis C virus (HCV) enzyme immunoassay (EIA) has resulted in an urgent need for well-characterized supplemental assays to confirm the presence of HCV antibodies. A comparative study of four commercially available supplemental assays is reported here: first- and second-generation versions of a strip recombinant immunoblot assay (RIBA-1 and RIBA-2), an HCV neutralization EIA, and HCV neutralization plus synthetic peptide EIA. Three hundred sixty-seven blood donor specimens that were repeatedly reactive on HCV EIA were studied. Most specimens (93%) were also evaluated by radioimmunoassay (RIA) with a six-antigen panel, and 60 selected specimens were tested for HCV RNA by the polymerase chain reaction (PCR). RIBA-1 and RIBA-2 gave concordant results with 86 percent of specimens, while an additional 13 percent were correctly classified by RIBA-2 but not RIBA-1. Neutralization EIA alone correctly identified 94 percent of the study group, while the remaining 6 percent required the peptide EIA or the combined neutralization-peptide assay system for correct classification. The RIBA-2 and neutralization-peptide assay system for correct classification. The RIBA-2 and neutralization-peptide assay systems yielded identical results for 86 percent of specimens, and these results were supported by RIA and selected PCR testing. Only 2 specimens (0.5%) were frankly discrepant, while 51 specimens were indeterminate on either (47) or both (4) assays. When either the RIBA-2 or neutralization-peptide assay yielded an indeterminate interpretation, the other system correctly classified the specimen (based on concordance with RIA and PCR data) in a high proportion (92%) of cases.(ABSTRACT TRUNCATED AT 250 WORDS)