OBJECTIVES: Incremental cost effectiveness ratios (ICERs) using quality adjusted life years (QALYs) are the main measure of cost-effectiveness used in decision making by health technology assessments (HTAs). However, there could be misinterpretation of these measures when used by non-health economists (e.g. clinicians and policy makers). Other health economic outcomes, such as number needed to treat (NNT) and cost of preventing an event (COPE) could provide a practical interpretation of the clinical and financial effectiveness of treatments for specific stakeholders. This study investigated how the use of NNT and COPE can support decision making.

METHODS: A literature review of publications and HTA submissions was conducted to determine where NNT and COPE could be used in decision making, and the appropriateness of existing guidelines available for these analyses.

RESULTS: The review identified several CEAs incorporating NNT and COPE analyses. These studies were primarily in treatments in which there are well defined events for the primary outcome measures, and where the QALY measure is less sensitive to the key clinical outcomes of interest. Cardiac and psychiatric disorders were the predominant areas in which NNT measures were utilized. The review identified little guidance from HTA bodies for these analyses and therefore further guidance is required to ensure the analyses conducted are robust.

CONCLUSIONS: This study shows NNT measures can be used in decision making, particularly with stakeholders who are less familiar with health economic methods. Generalisability, a key strength of the QALY measure, can be of less relevance for stakeholders interested in a specific disease area, therefore NNT and COPE can be used for treatment comparisons within a specific disease area. The main challenge in wider adoption of these approaches is a lack of standardisation across NNT models, and development of best practice guidelines would support wider use of these methods.