Background: In Colombia HIV/AIDS cases have increased over time exerting a greater financial burden on the government and healthcare providers for its management. For this reason, resources to treat HIV/AIDS population must be efficiently allocated to obtain the most benefits possible. Objective: This study compares the cost-effectiveness of 4 first-line HAART regimes for the treatment of HIV/AIDS over a 5-year time horizon, from a particular Colombian healthcare provider’s perspective. Methods: A Markov model was developed to simulate disease evolution through four health states based on CD4+ cell count. The initial distribution of the simulated population and new cases entering the system where based on information gathered from a reference cohort. For each health state, the model had input parameters of local direct medical costs and utility measures (QALYs). Sensitivity analysis for time horizon, probability of infection, and utility levels after treatment failures was performed. Results: Cost-effectiveness rankings show that 3TC/ZID+NEV has the lowest cost-effectiveness ratio (thousand USD$3.81 per QALY) and 3TC/ABC+EFV the greatest (thousand USD$4.77 per QALY). According to the ICER, only 3TC/ABC+EFV provides 2696 more QALYs at a cost of thousand USD$26.31 per QALY. Cost-effectiveness and incremental cost-effectiveness ratios were most sensitive to changes in the time horizon of the simulation. Yet, in all scenarios, the ranking by cost-effectiveness ratio remained constant. For longer simulation periods, the ICER is reduced below both thresholds of 3xGDP per capita and WHO-CHOICE for the region. Conclusion: All HAART therapies used are cost-effective based on the cost-effectiveness ratio, according to the ranking in the healthcare provider’s setting. 3TC/ABC+EFV can provide additional benefits for an additional cost. Thus, if the healthcare provider has enough resources, the differences in costs, should be assumed to guarantee the most benefits from HIV/AIDS