Objective: To investigate when hepatitis C virus (HCV) infection from mother to child occurs, and evaluate possible associafed factors. Design: Prospective c ohort study. Patients: Fifty four HCV infected children tested within three days of birth and their mothers. Main outcome measures: HCV RNA polymerase chain rea ction (PCR) results. Results: Seventeen of the children (31% , 95% confidence interval 19% to 46% )- were positive in the first 3 days of life and could be assumed to have acquired infection in utero. Testing PCR positive was not ass ociated with sex (53% v 49% boys; p = 0.77) or mode of delivery (29% elect ive caesarean section in both groups; p = 0.98). Children with evidence of intra uterine infection were significantly more likely to be of lower birthweight and infected with genotype 1 (58% v 12% , p = 0.01). Although a higher proportion of infants born to HCV/HIV co- infected women were PCR positive in the first 3 days of life, this difference did not reach statistical significance; excluding infants born to co- infected women did not affect the results. Thirty seven of the children (68% ) were negative in the first 3 days of life, 27 of whom were positive when tested again at 3 months, and nine were first PCR positive after 3 months (one child had no further tests). Conclusions: These results suggest th at at least one third and up to a half of infected children acquired infection i n utero. Although postpartum transmission cannot be excluded, these data suggest that it is rare. The role of HCV genotypes in the timing and mechanism of infec tion should be explored further. Objective: To investigate when hepatitis C virus (HCV) infection from mother to child occurs, and evaluate may associafed factors. Design: Prospective c ohort study. Patients: Fifty four HCV infected children tested within three days of birth and their mothers. Main outcome Results: Seventeen of the children (31%, 95% confidence interval 19% to 46%) - were positive in the first 3 days of life and could be assumed to have acquired infection in utero. Testing PCR positive was not associated with sex (53% v 49% boys; p = 0.77) or mode of delivery (29% elect caesarean section in both groups; p = 0.98). Children with evidence of intra uterine infection were significantly more likely to be lower birthweight and infected with genotype 1 (58% v 12%, p = 0.01). Although a higher proportion of infants born to HCV / HIV co-infected women were PCR positive in the first 3 days of life, this difference did not reach reach statistica Thirty seven of the children (68%) were negative in the first 3 days of life, 27 of whom were positive when tested again at 3 months, and nine were first PCR positive after 3 months (one child had no further tests). Conclusions: These results suggest th at at least one third and up to a half of infected children acquired infection in utero. Although postpartum transmission can not be excluded, these data suggest that it is rare. The role of HCV genotypes in the timing and mechanism of infec tion should be explored further.