During the last quarter of 1992, 984 patients from 13 dialysis centers in the Provence-Alpes-Côte-d'Azur region in France participated in a multicenter cross-sectional study to determine the prevalence, the risk factors, and the clinical consequences of infection by the hepatitis C virus (HCV). Serum samples were tested for anti-HCV antibodies using second-generation enzyme-linked immunosorbent assay (ELISA). In the case of a positive result, a combination test was performed using second-generation recombinant immunoblot (RIBA) or direct detection of HCV-RNA by nested polymerase chain reaction (PCR). Collected data included the patient's age, gender, cause of the kidney disease, type of dialysis treatment, number of years on dialysis, weekly dialysis time, drug addiction, co-infection with hepatitis B virus and human immunodeficiency virus (HIV), number of kidney transplants, number of blood transfusions, and history of acute or chronic hepatitis. Chronic HCV infection was detected in 232 (23.6%) patients, whereas only 71 (7.2%) were infected by HBV. Logistic-regression analysis showed that HCV infection was associated with dialysis over a long period, numerous blood transfusions, female gender, kidney grafts, HBV infection, hemodialysis, and acute as well as chronic hepatitis. Multiple-correspondence analysis confirmed that the contamination was both transfusional and nosocomial. These results underscore the need for a strict compliance with "universal precautions" (Centers for Disease Control [CDC], Atlanta) in dialysis units and raise the question as to whether anti-HCV-positive patients should be isolated.