Development and Validation of the TRansparent Uncertainty ASsessmenT (TRUST) Tool for Assessing Uncertainties in Health Economic Decision Models.

Abstract

BackgroundAn increasing number of technologies are obtaining marketing authorisation based on sparse evidence, which causes growing uncertainty and risk within health technology reimbursement decision making. To ensure that uncertainty is considered and addressed within health technology assessment (HTA) recommendations, uncertainties need to be identified, included in health economic models, and reported.ObjectiveOur objective was to develop the TRansparent Uncertainty ASsessmenT (TRUST) tool for systematically identifying, assessing, and reporting uncertainties in decision models, with the aim of making uncertainties and their impact on cost effectiveness more explicit and transparent.MethodsTRUST was developed by drawing on the uncertainty and risk assessment literature. To develop and validate this tool, we conducted HTA stakeholder discussion meetings and interviews and applied it in six real-world HTA case studies in the Netherlands and the UK.ResultsThe TRUST tool enables the identification and categorisation of uncertainty according to its source (transparency issues, methodology issues, and issues with evidence: imprecision, bias and indirectness, and unavailability) in each model aspect. The source of uncertainty determines the appropriate analysis. The impact of uncertainties on cost effectiveness is also assessed. Stakeholders found using the tool to be feasible and of value for transparent uncertainty assessment. TRUST can be used during model development and/or model review.ConclusionThe TRUST tool enables systematic identification, assessment, and reporting of uncertainties in health economic models and may contribute to more informed and transparent decision making in the face of uncertainty.Electronic supplementary materialThe online version of this article (10.1007/s40273-019-00855-9) contains supplementary material, which is available to authorized users.