OBJECTIVES: To estimate the cost-effectiveness of conjugated estrogens/bazedoxifene (CE/BZA) compared with conjugated estrogens/medroxyprogesterone (CE/MPA) or no treatment in a cohort of postmenopausal U.S. women exhibiting vasomotor symptoms (VMS), with an intact uterus and at least 12 months since their last menses.  METHODS: A Markov cohort model was developed to compare CE/BZA with CE/MPA or no treatment in terms of costs and Quality Adjusted Life Years (QALYs). The model considered VMS, vaginal bleeding, and eight long-term events (LTEs): Hip fracture, vertebral fracture, breast cancer, colorectal cancer, ovarian cancer, coronary heart disease, stroke and venous thromboembolic events. Health states were mutually exclusive. Women were followed through a VMS phase and an additional post-VMS phase in order to assess the treatment effects on the LTEs, representing a life-time horizon. Women were assumed to experience VMS and receive treatment for 4 years, additionally, onset of vaginal bleeding or any LTE resulted in treatment discontinuation. The model utilized data for treatment effects (estimated via network meta-analyses), risks, mortality, quality of life (QoL – using EQ5D estimates) and costs, which were identified from a literature review. The main outcome was cost per QALY. RESULTS: Results showed that CE/BZA lead to a gain in QALYs but was associated with higher total costs, resulting in a cost per QALY of $12,949 and $26,066 when compared with CE/MPA and no treatment, respectively. Driven by fewer bleeding events, women on CE/BZA received treatment for a longer duration than women receiving CE/MPA. The LTEs impacted results modestly. Uncertainty analyses indicated that results were robust to changes in key assumptions and input data, however results were most sensitive to changes in QoL associated with VMS and vaginal bleeding. CONCLUSIONS: CE/BZA is considered cost-effective for the treatment of VMS in postmenopausal women when compared with either CE/MPA or no treatment in the United States.