ABSTRACT
With the widespread use of modeling in pharmacoeconomics, sensitivity analysis has become an important tool for investigating the models being developed and used. In this respect, modeling is conceived as the simulation of complex systems in reality. In particular, a model can be defined as a simplification of such complex relationships, as simple as possible, yet reflecting all relevant aspects of reality. We know that such a model has to be both internally and externally valid, but, in addition, it is important to know its properties regarding the changes in the outcomes in relation to changes in the inputs or parameters. The set of parameters reflects those characteristics of reality that were deemed relevant for simulating the specific realities of interest. The latter may be the costs, savings, and health gains of a specific therapeutic treatment. The parameters may be concerning epidemiology, progression of disease, and unit costs. The generic investigation of these changes in the outcomes in relation to changes in the input parameters is generally labeled sensitivity analysis (SA).
