BACKGROUNDThe Australian National Herpes Zoster Immunisation Program commenced in November 2016 for people aged 70-79 years old in Australia but vaccine effectiveness (VE) in this setting has not previously been assessed.METHODSWe extracted records from two cohorts of patients aged 70-79 years in 2017 and 2018 respectively who were regular attenders in a nationwide general practice dataset, MedicineInsight. Cox proportional hazards models were used to estimate VE. Models were adjusted for potential confounders including age, sex, and other covariates. Analyses were also stratified by sex, presence of comorbid conditions and number of general practitioner (GP) visits in the previous year.RESULTSThe 2017 cohort included 40,275 regular attenders and the 2018 cohort 41,735. Both cohorts had a mean age of 73.9 years and 52% were women. In 2017, among vaccinated people, over 9,688 person-years of follow-up, 35 cases of zoster were diagnosed giving an incidence of 3.6 per 1000 person-years compared to 8.7 per 1000 person-years (264 cases/30,317 person-years) among unvaccinated people. For 2018, among vaccinated people there were 66 incident zoster cases over 16,716 person-years giving an incidence of 3.9 per 1000 person-years compared to 6.3 per 1000 person-years (156 cases/24,782 person-years) among the unvaccinated. Overall, in the first year of the program, when the average time since vaccination was about 8 months, VE was 63.5% (95% CI: 47.5, 74.6) but this fell to 48.2% (95% CI: 30.0, 61.7) in the second year when the average time since vaccination was about 18 months. We found no difference in VE across age, sex, presence of comorbid conditions, and prior GP visit frequency (P-interaction > 0.05).CONCLUSIONSVE was consistent with that estimated in other countries and international settings. However, our findings suggest waning effectiveness after the first year of the program. Further program evaluation is necessary.