OBJECTIVES: Azacitidine and decitabine are used to treat patients with myelodysplastic syndromes (MDS).  We sought to determine their cost-effectiveness.  METHODS:   We developed a Markov process model (1-month cycles) to track hypothetical cohorts of MDS patients treated with azacitidine or decitabine over 2 years. Model structure and parameters were derived from published literature, product labels, clinical trial data, and drug and medical services cost databases. Four health states were modeled: 1) MDS with transfusion dependence; 2) MDS with transfusion independence; 3) progression to acute myelogenous leukemia (AML); and 4) death. Cost-effectiveness was measured incrementally as: 1) cost per quality-adjusted life year (QALY); 2) cost per month of transfusion independence; and 3) cost per case of AML progression avoided.  The model used a third-party payer perspective with 2009 US costs.  One-way sensitivity analyses were performed on key model parameters.  RESULTS: The total number of QALYs (per 1000 patients) attained by azacitidine-treated patients exceeded those attained by decitabine-treated patients (1041 vs. 870). The total number of patient months with transfusion independence was higher for azacitidine vs. decitabine (8328 vs. 6224). More azacitidine-treated patients avoided progression to AML compared to decitabine-treated patients (509 vs. 285). Total per-patient costs for azacitidine were lower than for decitabine ($150,322 vs. $166,212). Overall, treating a patient with azacitidine cost $15,890 less than treating a patient with decitabine, and confers 0.171 additional QALYs.  CONCLUSIONS: These findings demonstrate that azacitidine costs less than decitabine and provides greater clinical benefit across key outcomes of interest. These results accentuate the role of azacitidine as a major asset in providing cost-effective care for MDS patients.