PCN268 - The Life and Death of the End of Life Treatment Appraisal Criteria in Nice Technology Appraisals?

Abstract

Since January 2009, NICE in the UK allows end of life (EOL) treatments to exceed the upper end (£30,000/QALY) of the threshold range for incremental cost-effectiveness ratios (ICERs) by using higher weights for EOL life-years. With discussions surrounding the concept and implementation, and the introduction of the value based assessment framework, the aim of this study was to review NICE technology appraisals (TAs) in oncology to assess the interpretation, implementation and implications of the criteria. All completed TAs in oncology since 2009 were searched. When multiple submissions of the same TA were made, the latest were selected. Data were extracted to capture details of the appraisal (e. g. treatment, indication and decision), the consideration of the five different EOL criteria (applicability, interpretation, effect on the decision) and the method of implementation (weighting, threshold). 61 TAs, including six multiple technology appraisals, assessing 71 technologies were reviewed. EOL criteria were considered in 40 TAs covering 44 technologies. EOL weighting was considered appropriate for 36% of technologies. Most technologies fulfilled the criterion of <24 month life expectancy (rejected in 14%), extension of life by ≥3 months or robustness of its calculation was the most common cause of rejection (32%/25% respectively). These criteria were inconsistently applied, using different methods (e. g. medians, restricted means from extrapolation, means from trial or model). The criterion of small population favoured technologies with limited indication (rejection 20%). Earlier TAs presented weight calculations, while later TAs only presented ICERs. Although aiming for greater transparency, the criteria left a large scope for interpretation in the decision making. Also, with the emphasis from applying higher weights to EOL life-years shifting to a differential threshold for certain indications, the original idea of considering wider societal preferences seem to have been neglected, that the new guidance should remedy.