We commenced to estimate the economic impact of salmeterol/fluticasone combination (SFC) therapy compared to fluticasone propionate (FP) therapy for asthma control in Japanese patients. A Markov model with five health states, developed by Price in 2002, was used. 1-week transition probabilities among status of asthma management were obtained from literature and epidemiological data from public data base. Direct cost for treatment was estimated from Japan medical fee schedule. Cost and effectiveness were not discounted due to 12-week simulation by the model. Univariate sensitivity analyses were undertaken to examine the main variables affecting cost-effectiveness. Probabilistic analysis was also undertaken to discuss statistical argument and to provide information for decision-making. In this analysis, the model was run over a 12-week period of time using transition probabilities. The results showed that treatment with SFC resulted in a higher proportion of totally controlled weeks per patient than treatment with FP (65.0 vs. 49.5%; incremental effectiveness by 15.5%), and lower mean direct asthma management costs ( yen168 702 vs. yen227 820). Probabilistic sensitivity analysis, conducted to assess robustness of the above base case result, showed that in the 95% of cases SFC was dominant (more effective and less costly) to FP. It suggested that SFC will be the most cost-effective therapy for asthma control. It would, however, be required to further evaluate cost-effectiveness of SFC in long-term observation.