OBJECTIVES: According to WHO, depression is a disease entity that generates affective disorders and causes severe impact on patient mood, which affects almost 121 million people worldwide. The aim of this analysis is to evaluate the cost-effectiveness of desvenlafaxine in major depressive disorder in the Colombian population from the institutional perspective. METHODS: A Markov model simulates the clinical course of a hypothetical cohort of patients >18 years with major depressive disorder. Patients may have up to three depressive episodes along the ten-state model developed through eight-week cycles, using a timeframe of five years and 3% annual discount rate. Comparators were: desvenlafaxine (50 mg/day), fluoxetine (20 mg/day), paroxetine (20 mg/day), duloxetine (60 mg/day) and escitalopram (20 mg/day). Effectiveness measures and adverse events were taken from a literature review, utilities and costs were taken from a local health institution (“Clínica Nuestra Señora de la Paz”), costs are expressed in 2012 USD. Effectiveness measures were depression-free weeks, days lost due to absenteeism, QALYs, and outcome was incremental cost-effectiveness ratio. RESULTS: Over a 5-year period, desvenlafaxine obtained 3.5129 QALYs, fluoxetine 3.4950; paroxetine 3.4896; duloxetina 3.4961, and escitalopram 3.4971. Total expected costs were for desvenlafaxine US$3640.30, fluoxetine US$3642.90; paroxetine US$3537.60, duloxetine US$3949.50 and escitalopram US$3275.30. Depression-free weeks were higher with desvenlafaxine (79.91) and lower with paroxetine (76.83); days lost due to absenteeism were lower with desvenlafaxine (69.19) and higher with paroxetine (383.28). CONCLUSIONS: For the treatment of major depressive disorder, desvenlafaxine is a cost-effective alternative compared to paroxetine (ICER = US$4,409.60/QALY) and escitalopram (ICER =US$23,104.50/QALY); desvenlafaxine is a cost-saving alternative compared to fluoxetina and duloxetina with a higher number of QALYs and lower costs.