OBJECTIVE: As an attenuated varicella vaccine is available, the question as to whether varicella could be eliminated in an acceptable time frame under cost-effective conditions in European countries arises. METHODS: Epidemiological data and results from modelling studies conducted in France, Italy, Spain, and Germany were analysed and compared. Input data for the German decision-analytic model were mainly taken from a seroprevalence study (4,602 sera) and a large epidemiological survey (1334 cases) in 1999. For France, an epidemiological model was developed and a prospective observational study of1832 cases was performed. The Spanish Markov model used data from a seroprevalence study, from hospitalisations, and from the available literature. The Italian figures were estimated from a one-year sentinel network in 1997. Costs were defined as costs of the vaccination program. Monetary benefits were defined as direct and indirect cost savings afforded by vaccination compared to a strategy not involving vaccination. Effectiveness was measured by the number of infections avoided. RESULTS: The age distribution of diagnosed cases in Italy and Germany were similar. The complication rates in Italy (3.5%) and France (2%) were lower than in Germany (5.7%). Indirect costs were the major contributors to disease cost and accounted for 67% of the cost in Spain and 80% in Germany. For all countries, vaccination programs dominated the non-vaccination strategy by saving costs for society and avoiding infections. Cost-benefit ratios varied between 1.61 (Spain) and 4.1 (Germany). The German model shows that with a coverage rate of 85% it would be possible to eliminate varicella within 15 years. CONCLUSION: Indirect costs turned out to be the most important cost factor. From an economic point of view, a universal vaccination program is justified. The most convenient and cost-effective means would be the implementation of a routine vaccination program using a combined measles, mumps, rubella and varicella vaccine.