A population-based case-control study was conducted in Taiwan to determine the hepatitis C virus (HCV)-associated risk of hepatocellular carcinoma (HCC) in a hyperendemic area for hepatitis B virus (HBV) infection. A total of 58 recently diagnosed HCC patients and 225 matched community controls, who participated in a community-based liver cancer screening program, were recruited between March 1991 and March 1994. Control subjects were matched to HCC patients by age (+/- 5 years), sex, residence, and date of serum sample collection (+/- 3 months). Serum samples from all study subjects were tested for hepatitis B surface antigen (HBsAg) and antibodies to HCV (anti-HCVs) by enzyme immunoassays, as well as HCV RNA by reverse transcription-PCR assays. Accordingly, patients with HCC were more likely than were controls to be positive for HBsAg (82.8% versus 12.9%, with an odds ratio (OR) of 22.9), anti-HCVs (13.8% versus 4.4%, with an OR of 3.9), and HCV RNA (13.8% versus 5.8%, with an OR of 2.7). After adjustment for anti-HCVs and HCV RNA positivities, the matched ORs associated with HBsAg increased to 27.6 and 28.1, respectively, whereas the corresponding adjusted ORs for anti-HCVs and HCV RNA after controlling for HBsAg status were increased to 8.8 and 6.2, respectively. In the meantime, interactive effects between HCV and HBV on risk were also observed. Compared with those who were negative for both anti-HCVs and HBsAg, the matched ORs associated with the sole positivity of anti-HCVs and HBsAg were 4.0 (95% confidence interval = 0.7-24.0) and 24.6 (95% confidence interval = 9.5-64.1), respectively, whereas 6 HCC cases but none of control subjects were positive for both anti-HCVs and HBsAg. These results indicate that there are mutual confounding and interactive effects between HCV and HBV with respect to their links to HCC in this endemic area of chronic HBV infections.