OBJECTIVES: A recent analysis of reimbursement recommendations issued by the NCPE in Ireland has identified quality of clinical evidence (QuEV) to be an important driver of reimbursement decisions; mirroring findings in other countries. The lack of a scoring system impedes the formal incorporation of this additional uncertainty in the decision making process. We propose a multidimensional scoring system (the QuEV score) to capture multiple aspects influencing quality of evidence. METHODS: Relevant dimensions for the assessment of QuEV were derived from the NCPE template to assess clinical and comparative effectiveness and complemented in discussion with the NCPE review team. The outcomes in each dimension are mapped onto values between 0 (=worst outcome) and 1 (=best outcome) and are combined in a analytic hierarchy process. This allows for continued flexiblity to represent preferences of different decision makers and a future incorporation into a MCDA model to inform reimbursement decisions. RESULTS: The following dimensions were included in the QuEV score (all to be evaluated with respect to the main comparator treatment): network properties (direct, indirect adjusted, indirect matched, naïve), trial type (multiple RCTs, one RCT, observational, phase-2 trials, phase-1 trials), trial quality (Cochrane risk of bias tool), and outcome measure (direct, surrogate). Three hypothetical treatment profiles illustrate how, analogously to actuarial science, the score can be used to translate into a threshold reduction. Sensitivity analysis on the relative importance of criteria highlights high sensitivity for treatments with varied scores in the dimensions. CONCLUSIONS: The QuEV score provides a quantitative tool to evaluate the underpinning clinical evidence in HTAs. Threshold reductions for technologies with an increased risk of a wrong decision due to a poor evidence base can ensure that the increased risk is not solely carried by the payer and provides an incentive for the pharmaceutical industry to collect post reimbursement data.