Objectives: Model-building is one of the most important phases of multiple-criteria decision analysis (MCDA). Defining the criteria is a critical stage in the process as it forms the foundation of MCDA. In order for the analysis to be meaningful, ultimately providing the highest possible insight to decision makers, criteria should adhere to a number of key properties otherwise results could be spurious. Our aim was to analyze the use of the different types of criteria in MCDA models in the decision-making process in healthcare, focusing on socioeconomics.

Methods: We reviewed scientific publications and official documents of existing or proposed MCDA models. We identified the criteria and assigned them into nine groups: Economics, Disease description, Intervention description, Health gain, Feasibility, Prevalence, Evidence quality, Social and ethical factors and Other criteria. We looked at MCDAs which are aimed to be used for Reimbursement or Investment Decisions (86%) and for Authorization Decisions (14%).

Results: Of the reviewed 25 MCDA models aimed to support reimbursement decisions only 1 lacks cost-related criteria, while 7 scoring systems lack social and ethical factors and 12 MCDAs don’t mention epidemiology or classification of target groups directly. It should be noted the necessity of the inclusion of economic factors is still debated.  For example, costs can be considered by incorporating the ‘impact on costs’ as a criterion, other than the total purchasing cost of the treatment itself or taking the incremental cost value ratio (ICVR) into consideration. We found that none of the reviewed MCDAs used for authorization decisions mentioned socioeconomic criteria.

Conclusions: It can be seen that while the selection of criteria has no standardized approach yet, in most existing or proposed MCDA models it has been constructed to adhere to the methodological frameworks and to be relevant for decision-makers in the particular decision problems.