In 1998/1999, an outbreak of mumps occurred among children of a religious community in North East London. A case control study was conducted to assess the effectiveness of the mumps component of the MMR vaccine. One hundred and sixty-one cases of mumps were identified and 192 controls were selected. Fifty-one percent of cases and 77% of controls had a history at least one MMR vaccination. The observed effectiveness of any MMR vaccination adjusted for age, sex and general practice was 69% (95% CI: 41-84%). This is consistent with the results of other observational studies of mumps containing vaccines, but lower than the immunogenicity of mumps vaccines reported by clinical trials. This discrepancy is because observational studies tend to underestimate vaccine effectiveness, and because immunogenicity is not necessarily an accurate biological marker of vaccine effectiveness. Two doses of vaccine were more effective (88% (95% CI: 62-96%)) than a single dose (64% (95% CI: 40-78%)). The current two-dose vaccination programme remains the best method for controlling mumps infection in the community.