Objective: To identify sensitive epidemiologic predictors of a positive hepatitis C virus antibody test in asymptomatic persons, and to compare the cost of testing only persons with an epidemiologic predictor to that of universal screening.Methods: Seventeen hundred consecutive pregnant women were tested by enzyme-linked immunosorbent assay for antibody to hepatitis C virus. Seventy-five subjects tested positive and were compared with 257 pregnant women who tested negative. Cohort and control patients were interviewed and their medical records were reviewed to identify those with chosen predictors of a positive hepatitis C virus antibody test.Results: Seventy-four of 75 cohort patients and 108 of 257 controls had one or more predictors of a positive antibody test. Cohort patients were significantly more likely (P < .001) to have the following: human immunodeficiency virus infection, a sex partner with a risk factor for hepatitis, age greater than 30 years, and a history of drug use, blood transfusion, sexually transmitted disease, hepatitis, or incarceration. The sensitivity and specificity of a single predictor in identifying a person with a positive test were 99 and 58%, respectively. The cost of finding a single individual with a positive antibody test by universal screening was $674, compared to $303 by selectively screening persons with one or more predictors of a positive antibody test.Conclusions: Most individuals with positive hepatitis C virus antibody tests can be identified on the basis of epidemiologic predictors, reducing the cost of testing by 55%. These patients may receive appropriate medical therapy, and their children may be evaluated for possible infection by vertical transmission of hepatitis C virus.