We have used a recently developed enzyme immunoassay to evaluate the prevalence of hepatitis C virus antibodies (anti-HCV) in patient groups and hospital personnel. The prevalence of anti-HCV in haemophiliacs, intravenous drug users, male homosexuals, and haemodialysis patients was 86, 63, 28, and 9 percent, respectively. Eight of 738 (1.1 percent) sera from health care workers were positive for anti-HCV. Only two of the 8 employees which were seropositive for anti-HCV also demonstrated serological markers of a previous infection with hepatitis B virus. The reported seroepidemiological data may not correctly reflect the risk of HCV transmission in the hospital setting, since the currently available assay shows a high sensitivity only for the detection of persistent HCV infections. There is an urgent need for the development of HCV antibody assays with increased sensitivity.