OBJECTIVE: There is ongoing discussion in the Netherlands about the introduction of hepatitis B (HepB) vaccination into the national immunisation programme. The purpose of this study was to analyse the cost-effectiveness of general vaccination of newborns in addition to the current policy, screening of pregnant women and subsequent immunisation of newborns from HBsAg-+ mothers (“screen + vaccinate”) in comparison with the current strategy of screening only (“screening only”), from a health care payer perspective. METHODS: The incidence of HepB infection over the next 50 years was simulated with a dynamic model, describing a population of age 0-60 years, with 6 sexual activity classes (both for hetero- and homosexual groups). Immigration from more endemic countries was modelled as yearly additional incidence of carriers in the population. Progress of disease after infection was computed with a Markov-chain model, that predicted future patient numbers, life-time costs of infection, and life-years lost after infection. The main outcome parameter was cost per life-years gained. Costs (in 1999 EUR) and effects were discounted at 4 %. RESULTS: Cost per life year gained appeared to be highly sensitive to the assumed percentage of carriers among immigrants, vaccination costs and to discounting of effects. Cost per life year gained in the screen + vaccinate scenario ranged from approximately EUR 25,000 (6 % carriers among immigrants) to EUR 75,000 (1 % carriers among immigrants). If effects were not discounted, cost per life year gained were below EUR 9,000 in all models explored. CONCLUSIONS: Compared to other interventions, hepatitis B vaccination is not cost-effective in low endemic countries such as the Netherlands. However, this conclusion depends heavily on the convention to discount effects at a similar rate as costs. The application of standard methodology is unfavourable for programmes that prevent negative health outcomes in the mid to long term.