OBJECTIVES: To evaluate the cost-effectiveness of oral anti-diabetic therapies, a fixed combination (metformin plus glibenglamide, Glucovance®) and a free combination (metformin plus rosiglitazone), in France. METHODS: A peer-reviewed, published, validated computer simulation model, was used to project long-term clinical and cost outcomes of treatment with Glucovance® or metformin+rosiglitazone. The model combined standard Markov sub-models to simulate the incidence and progression of complications and calculate costs over patients' lifetimes. Transition probabilities and risk adjustments were derived from published sources, including the UKPDS. Treatment effects, average doses and baseline cohort characteristics were taken from a recent double-blind randomized clinical trial showing that Glucovance® resulted in significantly greater reductions in HbA1c (-0.4%) compared with metformin+rosiglitazone in type 2 diabetes patients inadequately controlled on metformin monotherapy. Direct costs (2004 Euros) were retrieved from published sources and projected over patient lifetimes from a third party healthcare payer perspective. Costs and clinical benefits were discounted at 3% annually. RESULTS: In the long-term, Glucovance® treatment was associated with improvements in discounted life expectancy (0.19 years) and quality-adjusted life expectancy (0.18 years) compared to metformin+rosiglitazone. Undiscounted life expectancies were 14.73 and 14.40 years in the Glucovance® and metformin+rosiglitazone groups respectively. Diabetes-related complications were delayed by an average of 0.3 years in the Glucovance® arm. Lifetime direct costs (treatment and complication costs) were on average less expensive with Glucovance® than with metformin+rosiglitazone (by €5,605 per patient). At a shorter time horizon of 10 years, Glucovance® treatment was also projected to be associated with improved life expectancy and lower total costs than metformin+rosiglitazone. CONCLUSIONS: In France, Glucovance® is dominant to metformin+rosiglitazone over medium and long-term time horizons, leading to improvements in life expectancy, quality-adjusted life expectancy and cost savings in type 2 diabetes patients inadequately controlled on metformin monotherapy.