OBJECTIVES: It is considered best practice in economic modelling to account for adverse events (AEs) of new technologies, and this is required by many health technology assessment (HTA) agencies. The National Institute for Health and Care Excellence (NICE) recommends including costs and disutilities associated with AEs, however, it remains unclear how impactful AEs are on company-submitted analyses and final recommendations. This is particularly relevant for oncology medicines given the increasing number and complexity of appraisals.

METHODS: Guidance for 30 oncology NICE STAs published between 5^th October 2022 and 21^st June 2023 were identified. For each STA, the following information was reviewed:

• Methods of incorporating AEs within economic analyses.
• Impact of AEs on economic analyses through inspection of reported tornado diagrams and scenario analyses.
• Cost-effectiveness issues highlighted in NICE committee slides and guidance documentation.

Information relating to the impact of AEs on cost-effectiveness results was extracted and subsequently summarised.

RESULTS: Of the 30 STAs identified, 80% captured AEs through application of one-off costs/disutilities. 13 submissions included AE-related scenarios, with these negligibly impacting results. 16 tornado diagrams were available, of which one reported AE-related variables within the top 5 most-impactful variables. No committee slides or guidance documents reported AE-related assumptions as being key drivers of cost-effectiveness.

CONCLUSIONS: Findings indicate a limited AE impact on cost-effectiveness results and overall decision making. While there is a preference among companies for using one-off AE costs/disutilities, a range of approaches were reported, highlighting the lack of consensus on a single methodology. For NICE decision making, no guidance documents highlighted AEs as cost-effectiveness drivers. Therefore, adopting simplified approaches to capturing AEs may be most appropriate, particularly if more complex methodologies do not impact overall decision making. Further research is warranted to analyse whether this pattern persists within a larger sample size, and identify types of appraisals where AEs hold greater impact.