OBJECTIVE: The purpose of this study is to develop cost-effectiveness methodology in the context of a simultaneous modeling framework that provides consistent point and interval estimations. METHODS: A system of cost and effectiveness equations is suggested to model the simultaneity of the underlying cost and effectiveness variables, directly producing a marginal-effect measure of the incremental cost-effectiveness ratio of two competing medical interventions in an evaluative study. Empirical estimation of the simultaneous cost-effectiveness model was conducted using a feasible non-linear least squares estimation method. A simulation analysis of hypothetical data was performed to show the superior performance of the marginal-effect approach, relative to the traditional average-effect approach. RESULTS: Traditional average-effect approach has two shortcomings. First, it assumes two strong conditions: truly random distributions of all the significant non-intervention variables (both observed and unobserved) across a study’s intervention and control groups, and the independence of cost and effectiveness variables. Second, it does not give confidence interval, an important measure to assess the stochastic nature and robustness of point estimate. In contrast, the simultaneous marginal-effect approach imposed no restrictions on the randomness of the across-group distributions of all the variables. Furthermore, it takes into account the simultaneity of cost and effectiveness functions in estimation. The simulation analysis showed that the marginal-effect approach is significantly more robust, efficient, and unbiased than the average-effect approach in predicting the population true parameters assumed. CONCLUSION: The simultaneous marginal-effect approach should be chosen over the conventional average-effect approach whenever data allows in assessing the cost-effectiveness of competing interventions in medical decision making.