Health care workers (HCW) are a group at risk for hepatitis B virus (HBV) infection; as a result, vaccination is recommended. However, elevated cost of the vaccination schedule is one of the limiting factors to this approach. Our aim in this study was to evaluate alternative schedules for vaccination against hepatitis B, in order to obtain safe immunization with reduced costs. We studied 300 HCW, randomized to be submitted to one of three vaccination schedules against hepatitis B: Group A--three doses of 20 micrograms i.m. (n = 103); Group B--first dose i.d. (2 micrograms), second and third doses i.m. (20 micrograms) (n = 97); Group C--first and second doses i.d. (2 micrograms), third dose i.m. (20 micrograms) (n = 100). All individuals received recombinant vaccine at 0, 1 and 6 months. After the first dose, there was no difference among the three schedules, either in terms of anamnestic response or in seroconversion rate. After the second dose, there was statistical difference among the three schedules (A > B > C), in terms of seroconversion rates. After the third dose, seroconversion rates were 92.2% in Group A and 92.8% in group B; geometric mean titers (GMT) in Group B (789.6 UI l-1) were similar to group A (1248.0 UI l-1). Group C presented a seroconversion rate of 78% and a GMT of 323.0 UI l-1, both statistically inferior to other schedules. We concluded that the schedule applied in Group B had similar results when compared to schedule A, with estimated savings of 30% in vaccine costs.(ABSTRACT TRUNCATED AT 250 WORDS)