BACKGROUND AND AIM: Forty percent of new hepatitis B virus (HBV) infections in Australia occur in people who inject drugs (PWID); long-term infection carries the risk of serious liver disease. HBV incidence among Australian PWID has not been measured since the advent of targeted (2001) and adolescent school-based "catch-up" (1998) vaccination programs. We measured HBV incidence and prevalence in a cohort of PWID in Melbourne, Australia and examined demographic and behavioral correlates of exposure and vaccination. METHODS: Community-recruited PWID were surveyed about blood-borne virus risk behaviors and their sera tested for HBV markers approximately three-monthly over three years. Incidence was assessed using prospectively collected data. A cross-sectional design was used to examine prevalence of HBV exposure and vaccination at baseline. Poisson regression was used to identify correlates of HBV exposure and vaccination. RESULTS: At baseline, 33.1% of participants (114/344) had been vaccinated against HBV, 40.4% (139/344) had been exposed (previously or currently infected), and 26.5% (91/344) were susceptible. HBV incidence was 15.7 per 100 person-years. Independent associations with HBV exposure included female gender, South-East Asian ethnicity, drug treatment in the past three months, injecting in prison, and prior exposure to hepatitis C virus. Independent associations with vaccination included being ≤ 25 years old, reporting HBV vaccination, and never having been to prison. CONCLUSIONS: HBV infection continues at high incidence among Australian PWID despite the introduction of free vaccination programs. Innovative methods are needed to encourage PWID to complete HBV vaccination.