in order to observe the efficacy or a booster dose of hepatitis B vaccine, ic4 primaryschool children with a good response were enrolled in a double-bind, placebo-controlled and randomized field trial three years after the primary vaccination. At the end of the 6-year follow-up anti-HBspositive rate and GMT (of S/N value) In the revaccinated group were 54. 5% and 12. 0. still higherthan those in the control group (40. 5 % & 4. 8), but the difference of the positive rates was not statistically significant this time. Anti-HBs I,osltlve rate and GMT not only in the control group but inthe revaccinated group had dramatically declined against those 3 years arter the revaccination (thecontrol group: 69. 4 % & 20. 6 1 the revaccinated group: 87’ 8% & 43. 3) (P < 0. of ). The higher anti-HBs titer before the revaccination, the better the persistence of anti-HBs after the revaccination.HBV infection rate (calculated by person-year) In the revacclnated group was l’ 44%, without statistical difference from 3. 19% In the control group (P > 0. 05) as before. Considering the perfect longterm efficacy of hepatitis B vaccine, we concluded that a booster dose 9 years arter the primary immunization seems unnecessary. in order to observe the efficacy or a booster dose of hepatitis B vaccine, ic4 primaryschool children with a good response were enrolled in a double-bind, placebo-controlled and randomized field trial three years after the primary vaccination. At the end of the 6 -year follow-up anti-HBspositive rate and GMT (of S / N value) In the revaccinated group were 54. 5% and 12. 0. still higherthan those in the control group (40.5% & 4. 8), but the difference of the positive rates was not significant in this time. Anti-HBs I, osltlve rate and GMT not only in the control group but inthe revaccinated group had dramatically declined for those 3 years arter the revaccination (the control group: 69. 4 % & 20. 6 1 the revaccinated group: 87 ’8% & 43. 3) (P <0. of). The higher anti-HBs titer before the revaccination, the better the persistence of anti-HBs after the revaccination. HBV infection rate (calculated by person-year) In the revaccinated group was l ’44%, without statistical diffe rence from 3. 19% In the control group (P> 0.05) as before. Considering the perfect longterm efficacy of hepatitis B vaccine, we concluded that a booster dose 9 years arter the primary immunization seems unnecessary.