OBJECTIVE: To compare the incidence of adverse events following the administration of three commercially available measles-mumps-rubella (MMR) combination vaccines. METHODS: A randomized double-blind clinical trial was performed in 1996 that involved a total of 10 142 students 6-12 years of age in the state of Rio Grande do Sul, in Brazil. An MMR vaccine containing the Edmonston-Zagreb, Leningrad-Zagreb, and RA 27/3 strains ("vaccine A") was administered to 2 226 students (21.9% of the total); an MMR vaccine with the Moraten, Jeryl Lynn, and Wistar 27/3 strains ("vaccine B") was administered to 2 216 children (21.8%); and an MMR vaccine containing the Schwartz, Urabe AM-9, and Wistar 27/3 strains ("vaccine C") was given to 2 179 students (21.5%). A control group of 3 521 students (34.7%) was not vaccinated. Both the vaccinated subjects and the control subjects were followed daily for 30 days to detect any clinical manifestations. RESULTS: Adverse events were more frequent in the vaccinated children than in the control group (P < 0.01). In terms of causing parotitis, vaccine A had a relative risk (RR) of 5.72 (95% confidence interval (CI) = 3.11-10.54) when compared with vaccine B, and an RR of 2.33 (95% CI = 1.52-3.58) when compared with vaccine C. Vaccine A was also associated with an increased risk of lymphadenopathy when compared with vaccine B (RR = 3.11; 95% CI = 1.78-5.45) and with vaccine C (RR = 2.22; 95% CI = 1.35-3.66). Vaccine C was associated with an increased risk of parotitis when compared with vaccine B (RR = 2.46; 95% CI = 1.26-4.80). Three cases of aseptic meningitis were detected among the children in the study group, but only one case of vaccine-related aseptic meningitis was identified, among the children receiving vaccine A. CONCLUSIONS: The three MMR vaccines that we studied are associated with different risks of adverse events. We found vaccine A to cause more reactions than the two other vaccines, especially vaccine B. In addition, vaccine A presented both a temporal and a cause-and-effect association with one case of aseptic meningitis. We hope that this study will contribute information that can be used in choosing MMR vaccines with safe and effective strains, especially for mass vaccination strategies.