OBJECTIVES: To compare the cost-effectiveness of traditional treatment for ovarian stimulation in assisted reproduction techniques (ART) based on a combination of agonists of the Gonadotrophin released hormone (GnRH) versus a treatment based on the use of antagonist of the GnRH. METHODS: Cost-effectiveness analysis where both, effectiveness and costs are estimated by means of a clinical study carried out in the University Hospital Virgen de las Nieves of Granada (Spain) including patients in their first cycle. Effectiveness was measured as the number of ongoing pregnancies reached with each treatment. Direct costs were considered, for hospital and for patients as well. Markov model was used to simulate the clinical path of the compared techniques. Deterministic and probabilistic sensitivity analyses were used to analyse preliminary results.RESULTS: The clinical study included 274 patients, 138 treated with agonists and 136 with antagonists. No statistically significant differences were founded regarding age, cause of infertility and insemination technique used. Deterministic analysis showed the agonists’ treatment to be more effective than antagonists’ (33/138 versus 20/136 pregnancies respectively) but more costly (mean cost €3,204.37 versus €3,066.64€), with an incremental cost effectiveness ratio (ICER) of €14.96. Deterministic sensitivity analysis shows high sensitivity of ICER to effectiveness but not to cost of treatments. On the other hand, probabilistic analysis shows the agonist treatment to be a dominant strategy versus antagonist treatment. CONCLUSIONS: The use of antagonists of GnRH for ovarian stimulation in ART has been broadly discussed, but no cost-effectiveness studies have been carried out. Both, deterministic and probabilistic analyses show that, for women in the first cycle of ART, agonist treatment reports the best cost-effectiveness ratio in comparison with antagonist treatment. However, while the results of deterministic analysis are very sensitive to variations in treatment effectiveness, probabilistic analysis’ results reveal a robust dominance of agonists treatment.