RB51 vaccination can minimise the diagnostic problems associated with S19 vaccination of adult cattle, but its use for bovine brucellosis (BB) control remains controversial. Here, the evolution of BB prevalence in five high prevalence areas in Spain subjected to different control measures is described: herd depopulation of infected herds (I-III) or mass vaccination with RB51 and S19-RB51 vaccination of replacement heifers (IV-V). Annual data from the eradication campaigns were analysed at the special incidence area (SIA) level and the time to obtain herd prevalence levels of <1 per cent ('controlled status') was obtained at the local veterinary unit (LVU) level and compared using Cox's proportional hazard model. A higher annual rate of decrease in herd prevalence was observed in the SIAs subjected to vaccination (46.9%, 95% CI 43.5% to 50.0%) compared with those managed using stamping out (14.9%,95% CI 9.6% to 19.9%). No significant differences in the time to achieve controlled status were observed between the stamping-out and vaccination strategies used at the LVU level, with median times of 60 (stamping-out LVUs) and 63 (vaccination LVUs) months. These results suggest that RB51 mass vaccination, in combination with the S19-RB51 vaccination of replacement heifers and strict implementation of other eradication meas res, may provide results at least comparable with those resulting from a herd depopulation based strategy.