A randomized multicenter study compared the routine hepatitis B vaccine schedule of 0, 1, 6 months with an accelerated schedule of 0, 1, 2 months in newborns. Two hundred ninety-nine infants whose mothers were seronegative for hepatitis B were enrolled in the study and randomized to either the routine or accelerated schedule. All infants had blood drawn for antibody titers to hepatitis B at 2, 3, 6 and 7 months of age. For 222 infants data were evaluable, at least for safety; 193 of these 222 had antibody titers that were evaluable. The infants vaccinated on the accelerated schedule developed seroprotective concentrations of antibody more quickly than the infants vaccinated on the routine schedule; 92.6% vs. 66.1% had seroprotective concentrations (≥ 10 mIU/ml) at 3 months of age (P < 0.001). However, infants in the accelerated schedule had lower geometric mean antibody titers at 7 months, 420.0 vs. 3141.8. We conclude that the accelerated vaccination schedule resulted in the more rapid development of seroprotective concentrations of antibody, but levels of antibodies were not as high as in the routinely vaccinated infants at 7 months. These data suggest that an accelerated vaccine schedule can be used in the newborn period. The effectiveness of the accelerated schedule in preventing perinatal infections compared to the standard schedule and the necessity for booster doses of vaccine remain to be studied.