This trial evaluated the reactogenicity, kinetics of antibody induction, and long-term immunogenicity of a 720 enzyme-linked immunosorbent assay units (EL.U.) antigen dose of an inactivated hepatitis A vaccine (Havrix, SmithKline Beecham Biologicals, Rixensart, Belgium). One hundred six healthy adult volunteers were enrolled to receive vaccine intramuscularly according to a 0, 1, and 6-month schedule. The vaccine was well tolerated. The most frequently reported local symptom was soreness, observed following 37.1% of all doses. Headache was the most frequently reported general symptom observed following 12.9% of documented vaccine doses. The administration of one vaccine dose induced seropositivity (anti-hepatitis A virus [HAV] > or = 20 mIU/ml) in 91% of all vaccinees 1 month later. The second vaccine dose resulted in seropositivity of the remaining vaccinees at month 2. All subjects remained seropositive for HAV antibodies at month 6, at which time the booster vaccine dose was given. At month 7, all vaccinees had anti-HAV titres > 200 mIU/ml. Serological results obtained at months 12, 18, 24, and 36 showed that antibodies against HAV induced by the vaccine booster dose persist for at least 30 months following its administration. All 49 subjects followed up until month 36 had antibody titres > or = 20 mIU/ml. The geometric mean titre (GMT) decreased by 60% from month 7 to month 12; between month 12 and 36, the GMT decreased by approximately 14% per period of 12 months. According to the vaccine-induced antibody kinetics and the magnitude of antibody level decrease over time, the predicted duration of antibody persistence is estimated to be at least 20 years.