A new hepatitis C virus (HCV) core antigen (HCV Ag) assay was thought to have a good correlation with HCV RNA. The aim was to elucidate the usefulness of this HCV Ag assay in community screening. In a township where HCV is endemic, 405 residents aged 58 years or older responded to a follow-up community screening. All subjects were tested for anti-HCV (AxSYM, version 3.0; Abbott Diagnostics) and HCV Ag (Architect HCV Ag test; Abbott Diagnostics). For subjects with anti-HCV signal-to-cutoff ratios (S/CO) > 10 and/or HCV Ag > 3 fmol/liter, HCV RNA data (Taqman HCV RNA; Roche Diagnostics) were further checked. A total of 115 (28.4%) subjects had their serum HCV RNA levels measured, and 93 were HCV RNA positive. The other 290 subjects were supposed to be HCV RNA negative. HCV Ag was significantly correlated with HCV RNA according to the following equation: (log HCV RNA) = 2.08 + 1.03 (log HCV Ag) (R(2) = 0.94; P < 0.001). As determined using a combination of the values for anti-HCV (S/CO > 40) and HCV Ag (>3 fmol/liter) as a cutoff to predict viremia, the sensitivity, specificity, accuracy, positive predictive value, and negative predictive value were 96.8%, 100%, 99.3%, 100%, and 99%, respectively. In conclusion, for a community study, HCV Ag showed good correlation with HCV RNA. In addition, anti-HCV or HCV Ag can predict HCV viremia well, while a combination of anti-HCV (>40 S/CO) and HCV Ag (>3 fmol/liter) can provide the best result validity.