OBJECTIVES: In France, the CEESP (Economic & Public Health Assessment Commission) appraises the health technology assessment submitted by the manufacturers when requested concomitantly with a reimbursement demand. The CEESP establishes the methodological acceptability of the health economic assessment through methodological reservations qualified as minor, important or major.

Since 2021, the CEESP also specified the level of uncertainty surrounding the results, could be driven by the difficulty of estimating key parameters, by a lack of credibility of hypotheses or by the interpretability of the results.

A major reservation or uncertainty invalidates the dossier and the economic value that it defends.

Using the Vyoo Agency database, which collects all the opinions published by the HAS, the objective is to analyze the main reasons retained by the CEESP to invalidate the economic information among health economic appraisals published in 2023.

METHODS: All available Health Economics opinions published by CEESP between January 1^st and December 31^st, 2023, were reviewed.

RESULTS: In 2023, 25 appraisals were disclosed, 9 of which were invalidated. Eight major methodological reservations were stated, invalidating 6 dossiers. The major reservations relate to effectiveness modelling (4), utility estimation (3) and type of analysis. Relating to the effectiveness, all reservations concern the indirect comparisons. Two reservations relate to the robustness of the network meta-analysis and the others to the acceptability of the method used. Relating to utility estimation, 2 reservations relate to the use of a vignette study. In addition, 3 dossiers were invalidated due to major uncertainty mainly supported by the evolving context or statistical variability of the parameters.

CONCLUSIONS: In 2023, 64% of dossiers were likely to provide economic information useful in the decision-making process in France, compared with less than 50% in 2022. Increasing the number of validated assessments is a key issue to ensure economic information can be fully used in pricing negotiations.