Experimental hepatitis B immunization trial in newborns was carried out in Burundi. Newborns were randomly divided into vaccine and control groups. Vaccinated newborns were given two infections of hepatitis B vaccine: one at birth and another 2 months later. A booster dose was given at 12 months of age. Results obtained show that two months after the second dose of HB vaccine, 96.8% of the vaccinated babies had anti-HBs; at the age of one year this figure had fallen to 83.8%. Six months after the booster dose, 95.6% were anti-HBs positive, with a geometric mean titre of 214 mIU ml⁻¹. The anti-HBs responses in these infants was compared to those observed in previous studies performed in Senegal in infants (same protocol) or in newborns (3 dose protocol). The anti-HBs responses were lower in terms of mean titre values in neonates who received the two dose protocol than in older children and in neonates who received three doses at one month intervals. Vaccine efficacy was monitored during a two year period in neonates both immunized and nonimmunized. Protective efficacy was found to be 100% if considering HBsAg positive events and 75% if considering all HBV events (HBsAg and/or anti-HBc positive).