Background: Since 2022, clade IIb mpox outbreaks have occurred in non-endemic countries, primarily amongst men who have sex with men. The World Health Organization currently recommend two doses of modified vaccinia Ankara-Bavarian Nordic vaccine (MVA-BN) at least four weeks apart for people at risk of infection. Infections in fully vaccinated people have been infrequently reported. We aimed to determine whether vaccination was associated with attenuated disease. Methods: An outbreak investigation and retrospective cohort study of mpox cases notified in New South Wales, Australia between 20 June - 20 November 2024 was conducted. Cases and clinicians were interviewed. Cases were classified as fully vaccinated if they had received two pre-exposure MVA-BN doses and partially vaccinated if they had received one dose prior to exposure. Risk ratios were calculated for clinical outcomes by the number of MVA-BN doses. A subset of monkeypox genomes were sequenced. Findings: Sequencing of 102/674 (15.1%) cases showed they were all clade IIb. Vaccination status was ascertained for 98.4% of cases (663/674), with 37.2% fully vaccinated, 10.7% partially vaccinated, and 50.4% unvaccinated. The median time between vaccination and symptom onset was 21.8 months (IQR 19.5-23.0 months). Compared to unvaccinated cases, those who had previously received two MVA-BN doses were less likely to be hospitalised (RR 0.11 [95%CI 0.03-0.43]), have extragenital lesions (RR 0.45 [0.36-0.56]) or systemic symptoms (RR 0.72 [0.64-0.80]). Interpretation: Two MVA-BN doses reduced the risk of disseminated disease and hospitalisation, including in cases vaccinated >1.5 years ago.