Presentation (CTI)

OBJECTIVES: The National Institute for Health and Care Excellence (NICE) methods guide (2022) states that adverse events (AEs) should be considered in evidence submissions, where they materially affect costs or health outcomes. This targeted review explores how recent NICE technology appraisals (TAs) capture AE disutility values and costs by consolidating and discussing methodological approaches.
METHODS: NICE TA committee papers published over the last 5 years (from May 2021 to June 2026) were screened. Terminated, replaced, and withdrawn appraisals were excluded from further analysis. All remaining appraisals were classified as ‘cost eligible’, as AE resolution costs could be considered across all appraisal types. Appraisals were also classified as ‘disutility eligible’, where AE disutilities could be considered, e.g. cost comparison appraisals were not considered ‘disutility eligible’. Where AE costs or disutilities were not reported, the rationale for omission was recorded.
RESULTS: Of the 464 appraisals screened, 124 were excluded due to being replaced (n=21), terminated (n=98), or withdrawn (n=5). Of those remaining, 305 were deemed ‘disutility eligible’, and 340 were ‘cost eligible’. Only 194 (64%) disutility-eligible appraisals reported AE disutility values, whilst 246 (72%) cost-eligible appraisals reported AE resolution costs. Oncology was the most frequently assessed disease area with 168 (55%) disutility-eligible and 176 (52%) cost-eligible appraisals. Of these, 125 (74%) reported AE disutility values and 155 (88%) reported AE resolution costs. Other frequently reported disease areas, such as autoimmune diseases, included AE disutility values and costs only 50% of the time. Reasons for exclusion were not reported consistently.
CONCLUSIONS: NICE TAs had more complete reporting for AE resolution costs in comparison to disutilities. Reporting across disease areas was variable, reflecting differences in the number of appraisals across disease categories. Inconsistencies in AE reporting (generally, and across disease areas) highlight the need for clearer methodological guidance to support more standardised approaches to model AEs.