Incorporation of statistical uncertainty in health economic modelling studies using second-order Monte Carlo simulations.

Health economic modelling studies are of interest to many parties with different responsibilities and diverging interests. Therefore, it is obvious that recognising the relevance of statistical uncertainty and dealing with it appropriately are required to obtain unbiased results from health economic modelling studies, especially when those data are being used for reimbursement decisions. In this manuscript we explore the relevance of the incorporation of statistical uncertainty in a health economic model and identify various types of statistical uncertainty. The concepts were applied to a hypothetical Markov model for a hypothetical antiparkinsonian (AP) product. The method was based on the incorporation of probability distributions in the input variables using a second-order Monte Carlo simulation and the definition of minimum relevant differences for clinical and economic input variables and outcomes. Our paper shows that the outcomes of a health economic model might be severely biased when statistical uncertainty is not taken into account, which justifies the need for the incorporation of statistical uncertainty in a health economic model.