OBJECTIVES: Cost-effectiveness analyses (CEA) have become widely used in several pathologies.  Currently, new CEA studies comparing active vs control treatment have been incorporated each year.  For this reason, the combination of CEA studies could give a more consistent and accurate estimate of an intervention’s efficiency than one study alone. The aim of study was to develop a new method to do comparative efficiency research (COMER) based in individual patient data. METHODS: After, adjusted the marginal distribution and copula distribution of a hypothetical cohort, we stated the parameters and distribution estimated like our unknown theoretical distribution.  We conducted an iterative analysis of a random Frank Copula distribution with a different range of sample size. We performed a comparison between samples and theoretical distribution in terms of incremental cost-effectiveness ratio (ICER), incremental monetary benefit(IMB) fixed a threshold (k= 20.000 monetary units) and goodness of fit for Frank copula, assuming a tolerance. RESULTS: The Theoretical distribution fixed, showed a cost of 604.34 monetary units for active and 516.12 monetary units for control, and a utilities of 0.529 for active and 0.492 for control. ICER for theoretical cohort was 2,380 monetary units per quality-adjusted life year gained and IMB was 653. With a tolerance of 500 monetary units for ICER and 50 monetary units for IMB, only 15.52% of simulations were near the theoretical ICER and only 6.12% of IMB. The amount of individual patients simulated was more than 500 patients per treatment to fit Frank Copula. CONCLUSIONS: Preliminary results showed that COMER based in individual patients’ data could allow decision maker to know real add-value of a new intervention.